Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Disorders in sodium and water homeostasis primarily result from perturbation in the release or response to arginine vasopressin (AVP). Impaired AVP secretion or response results in impaired renal concentration and is termed diabetes insipidus (DI). Hyponatremia that results from AVP production in the absence of an osmotic or hemodynamic stimulus is termed syndrome of inappropriate antidiuretic hormone secretion (SIADH). The goals of this chapter are to review both DI and SIADH, their causes, and their treatment.
The plasma osmolality is maintained within a narrow range, 275–290 mOsm/L. The hypothalamus senses alterations in osmolality which leads to changes in thirst and AVP release to return the osmolality to normal. Osmoreceptors are located in the anteroventral third ventricle of the hypothalamus and AVP is synthesized in the vasopressinergic neurons of supraoptic nuclei and paraventricular nuclei of the hypothalamus. AVP is stored and subsequently released from the posterior pituitary ( Fig. 8.1 ). The gene that encodes AVP is located on chromosome 20. The biosynthesis of AVP proceeds by way of a polypeptide precursor. This prohormone contains the AVP moiety, a protein known as neurophysin II (NPII) and a glycoprotein. This prohormone ultimately folds and dimerizes and is cleaved into AVP, where it is stored in the posterior pituitary until it is released into the cavernous sinus and superior vena cava.
AVP is the primary determinant of free water excretion. Its major renal effect is to increase water permeability in the medullary and collecting tubules via the insertion of the aquaporin-2 (AQP2) water channel on the apical surface of the cortical collecting duct ( Fig. 8.2 ). The action of AVP is mediated via the vasopressin V2 receptor that is primarily located on the basolateral surface of the principal cells. The V2 receptor is coupled to a guanine-nucleotide-binding protein, Gs. AVP binding of V2 receptors results in activation of adenylate cyclase with increases in cAMP and activation of cAMP-dependent protein kinase A (PKA). Activated PKA phosphorylates the serine residue at the C-terminus of the AQP2 protein which then leads to insertion of AQP2 on apical surface of the principal cells of the cortical collecting duct which markedly increases water permeability. Water movement across the basolateral membrane is facilitated by AQP3 and AQP4. Any defect along this pathway will prevent AQP2 expression and will result in a renal concentrating defect. Any process that will increase the expression of AQP2 will result in an impaired ability to excrete free water.
DI is a polyuric condition due to either inadequate vasopressin production, termed central DI, or the renal resistance to vasopressin, termed nephrogenic DI. An unusual form of DI, gestational DI, occurs during pregnancy from the degradation of plasma AVP by excessive vasopressinase produced in the placenta. This condition responds to the administration of desmopressin (dDAVP) as it is not degraded by vasopressinase and remits following delivery. Polyuria, unrelated to a disorder in AVP production or response, can be due to excess fluid intake. This condition is referred to as primary polydypsia and can be seen in compulsive water drinking or psychogenic polydypsia. Primary polydypsia can be difficult to distinguish from DI at times. Polyuria is typically defined as the passage of large volumes of hypotonic urine with an osmolality less than 300 mOsm/kg and in excess of 3 L/1.73 m 2 per 24 h.
Nephrogenic diabetes insipidus (NDI) refers to polyuric states despite the adequate production of AVP. NDI can be congenital or acquired, associated with severe polyuria or mild nocturia. Many pathophysiological mechanisms can result in a state of NDI:
resistance of AVP in the collecting tubule of the kidney
impairment in the medullary countercurrent mechanism
increase solute excretion by the kidney.
Genetic |
|
Sickle cell disease |
Electrolyte abnormalities |
|
Medications |
|
Acquired forms of NDI are the most common encountered in clinical practice, but the congenital forms are the most severe. The presenting features of NDI are usually polyuria or nocturia. Hypernatremia will only be a presenting feature if there is an impairment in access to water, such as during an acute illness or hospitalization. Numerous medications and disease states can result in an acquired NDI, so a high index of suspicion is required to make this diagnosis when a patient presents with or develops polyuria or hypernatremia.
Congenital nephrogenic diabetes insipidus (CNDI) is a rare hereditary disorder with an estimated prevalence of 1 in 250,000 males. There are two forms of inheritance, an X-linked recessive form, which affects the majority of individuals, and autosomal recessive and dominant forms, which affect the minority. Individuals can be equally affected with either form. In CNDI, there are varying degrees of resistance to both exogenous and endogenous AVP in the collecting tubules of the kidney. There is an inability to appropriately reabsorb free water in the collecting duct, but urinary electrolyte handling is normal. First described in 1945, the cardinal features of this condition are polyuria, polydypsia, fever, and episodes of hypernatremic dehydration. Pregnancies can be complicated by polyhydramnios, though this is not usually the case. Approximately 50% of affected males develop symptoms soon after birth. Affected infants have severe polyuria and polydypsia, a penchant for drinking cold beverages and frequently have constipation from dehydration. Complications reported in CNDI include mental retardation from repeated episodes of hypernatremia, intracranial calcifications, failure to thrive from inadequate caloric intake, and nonobstructive hydronephrosis from the high urinary flow rate. Patients should have periodic renal sonograms as severe hydronephrosis has been reported. In general, patients who have been diagnosed early and received treatment with good supportive care achieve a good quality of life without serious long-term complications.
CNDI can result from either loss-of-function mutations in the arginine vasopressin receptor 2 (AVP2) or the aquaporin 2 (AQP2) gene. The AVP2 gene is located in the region of chromosome Xq28. Defects in AVP2 gene are X-linked recessive and affect approximately 90% of patients with CNDI. There have been almost 200 putative gene mutations in AVPR2 resulting in X-linked CNDI. Males are severely affected and do not increase their urine osmolality in response to exogenous vasopressin. Males with X-linked CNDI typically have a maximal urine osmolality of less than 300 mOsm/kg per H 2 O. A severely affected male can produce over 10 L of urine a day. Females have varying degrees of renal concentrating defects. Many females are asymptomatic. The AQP2 gene is located in the region chromosome 12q3. Mutations in the AQP2 gene can be either autosomal recessive or dominant and affect approximately 10% of families with CNDI, with 90% of these being autosomal recessive. There are approximately 30 putative AQP2 gene mutations. A diagnosis of CNDI can usually be made on history, as most children present in the first year of life, but molecular genetic testing and prenatal testing is clinically available.
The primary management of CNDI is adequate administration of free water to replace urinary losses. Patients with CNDI should be on a low sodium diet to decrease the renal solute load. Infants should be on a low solute infant formula and offered fluid approximately every 2 h. Infants typically prefer their formula at room temperature or slightly chilled while older children and adults crave ice water. A variety of medications have proven successful in the management of CNDI including hydrochlorothiazide, amiloride, the prostaglandin synthetase inhibitor indomethacin, and COX-2 inhibitors. A combination of hydrochlorothiazide and amiloride is most often recommended by experts in the field.
Studies have demonstrated that thiazides can decrease urine volume by as much as 50% and almost double the urine osmolality in patients with DI. It is not fully understood how thiazide diuretics induce this paradoxical effect. The most widely held view is that thiazides induce extracellular volume contraction via increased renal sodium excretion, which results in a slight decrease in GFR and increase proximal tubular sodium and water reabsorption. More recent data have emerged that thiazide diuretics directly increase the water permeability in the inner medullary collecting ducts and may upregulate the expression AQP2, NaCl cotransporter, and the epithelial sodium channel. The addition of amiloride to hydrochlorothiaze has an additive affect in reducing urine output and also helps prevent the thiazide-induced hypokalemia.
Nonsteroidal antiinflammatory drugs (NSAIDs) and COX-2 inhibitors have been used successfully in patients with CNDI, but their clinical use has been limited by the undesirable side effects associated with these medications, such as gastrointestinal (GI) bleeding, renal toxicity, hematopoietic effects, and cardiac ischemia. The exact mechanism of prostaglandin inhibitors is uncertain. Prostaglandins are known to diminish the effect of AVP by inhibiting adenylate cyclase. For this reason, NSAIDs are believed to potentiate the effects of AVP. The effects of NSAIDs may also be due to reduced medullary blood flow, causing a further reduction in glomerular filtration rate (GFR) with increased proximal tubular reabsorption of sodium. Not all NSAIDs are effective in the treatment of CNDI. Indomethacin has been the most effective and ibuprofen has not been effective.
The exogenous administration of vasopressin is generally not effective in the treatment of CNDI. There may be a subpopulation of patients that partially respond and this could be of clinical benefit. Surprisingly, the exogenous administration of dDAVP has been found useful in the treatment of nocturnal enuresis via extrarenal effects of vasopressin that convert enuresis to nocturia.
Acquired nephrogenic diabetes insipidus (ANDI) is much more common than CNDI and is also less severe. In most forms of ANDI, there is only a partial renal concentrating defect. Patients can be asymptomatic or have polyuria, polydypsia, and nocturia. Hypernatremia is less common, but severe cases have been reported with intercurrent illnesses. ANDI can be either chronic in nature when occurring in the context of renal disease or reversible when due to medications or electrolyte abnormalities. ANDI should be suspected in:
patients with polyuria, urine output > 3 L/1.73 m 2 per day
a urine osmolality < 300 mOsm/kg per H O or a urine specific gravity < 1.015 on multiple occasions
mild hyperosmolality, serum osmolality > 295 mOsm/kg per H O, or a serum sodium > 143 mEq/L, under normal conditions of fluid intake
a less then maximally concentrated urine < 800 mOsm/kg per H O in the face of hypernatremia.
Renal concentrating defects have been observed in a variety of renal diseases. There are many possible explanations for renal concentrating defects including decreased tubular responsiveness to AVP, damage to the renal medullary interstitium or impairment of the countercurrent mechanism, and increased solute excretion in the remaining functioning nephrons. The degree of renal concentrating impairment parallels the decline in GFR, with renal isosthenuria or hyposthenuria being an almost universal finding in advanced chronic renal failure. One of the primary mechanisms of renal concentrating impairment is vasopressin resistance, explaining the hyposthenuria in renal disease. Chronic renal failure induced by 5/6 nephrectomy in rats results in polyuria due to decreased expression of AQP2 and AQP3. Polyuria is a common complication following ureteral obstruction. Rat models of bilateral ureteral obstruction have revealed markedly reduced expression of AQP2. AQP2 expression remains low after the ureteral obstruction has resolved and renal concentrating defects can still be documented as much as 1 month postobstruction. Polyuria is also a common feature in the recovery phase of ischemic acute renal failure. The mechanism behind this appears to be both defects in countercurrent multiplication and impaired water permeability in the collecting duct due to decreased expression of AQP2 and AQP3.
Renal concentrating defects are a universal finding in patients with sickle cell anemia and sickle cell trait. The extent of the renal concentrating defect is largely dependent on the percentage of hemoglobin S. The renal medulla is an environment conducive to hemoglobin S polymerization and red cell sickling as there is high rate of oxygen consumption with hypoxia, an acid environment, and hypertonicity. Red cell sickling results in vaso-occlusion within the vasa recta with consequent inner medullary and papillary damage. Microradioangiographic studies have revealed almost complete loss of vasa recta in kidneys from patients with sickle cell disease. The renal concentrating defects primarily results from the loss of deep juxtaglomerular nephrons that are necessary for maximal renal concentration. The renal concentrating defect can be reversible early on with red cell transfusions but in adulthood is irreversible. Exogenous vasopressin and indomethacin have little or no effect in improving renal concentration. Patients with sickle cell disease usually have little difficulty maintaining water homeostasis unless there has been significant volume depletion or water deprivation.
Lithium is the most common cause of drug-induced NDI causing a renal concentrating defect in approximately 55% of patients on chronic lithium therapy, with overt polyuria, >3 L/day, occurring in 20%. Severe life-threatening hypernatremia has been reported to occur, especially in elderly patients with restricted access to water. Lithium is excreted mainly by the kidney and is reabsorbed in competition with sodium. NDI poses a risk to patients on lithium therapy as volume depletion can promote lithium retention via activation of the renin–angiotensin–aldosterone system, leading to increased lithium absorption and acute lithium toxicity precipitating a viscous cycle of worsening NDI. Decreased dietary sodium alone can lead to lithium toxicity. The degree of renal impairment is dependent on both the duration of lithium therapy, average serum lithium ion level and the cumulative dose of lithium carbonate administered. Even in the well-controlled patient, urinary lithium levels are sufficiently high to impair renal concentration. The renal concentrating defect caused by lithium is not always reversible and permanent renal concentrating defects have been reported after lithium discontinuation. Amiloride has been used successfully in treating lithium-induced NDI. Amiloride appears to decrease the uptake of lithium in the principal cells of the collecting duct, thereby preventing the inhibitory of effect of lithium on water transport.
How lithium actually causes DI is complex and appears to involve multiple mechanisms. Initial animal studies revealed that lithium impaired water handling via decreased adenylate cyclase activity resulting in decreased generation and accumulation of cAMP in response to AVP. Lithium appears directly to inhibit the activation of vasopressin-sensitive adenylate cyclase in renal epithelia by competing with magnesium for activation of the GTP-binding protein, Gs. Experiments in rat kidney medulla reveal that lithium caused marked downregulation of AQP2 expression that is only partially reversed by cessation of lithium therapy. In human studies, lithium exposure results in decreased urinary excretion of both AQP2 and cAMP. Recent experiments in mouse cortical collecting duct cell lines have revealed that short-term lithium exposure can decrease the expression of AQP2 via the AQP2 mRNA and independent of cAMP and adenyl cyclase. Chronic lithium exposure can also affect sodium and water handling by changing the cellular profile in the collecting duct. Chronic lithium exposure decreases the percentage of principal cells in the rat cortical collecting duct, which are responsible for hormonally regulated sodium and water handling, and increases the percentage of intercalated cells that are involved in acid–base balance. The cellular architecture can improve following the discontinuation of lithium. An additional mechanism for the polyuria seen with lithium administration is that of increased urinary sodium excretion. This appears to be due to a dysregulation of the epithelial sodium channel through a decreased responsiveness to vasopressin and aldosterone in the cortical collecting duct.
Total body potassium depletion can result in a significant renal concentrating defect in humans. Human subjects with a potassium deficit exceeding 300 mEq are only able to concentrate their urine to approximately 300 mOsm/L and some individuals develop hyposthenuria. The renal concentrating defect is a consequence of chronic potassium depletion and not acute hypokalemia. It has been demonstrated that rats fed potassium-free diets manifest renal concentrating defects before the development of hypokalemia. The primary cause of the concentrating defect appears to be decreased expression of AQP2, but the mechanism behind this is unclear. Chronic hypokalemia results in multiple morphological changes in the kidney, so there may be multiple factors that produce polyuria.
Impaired renal concentrating ability is a universal finding in patients with hypercalcemia and hypercalciuria, regardless of the etiology. A renal concentrating defect typically occurs when the serum calcium exceeds 11 mg/dL. Patients with serum calcium greater than 11 mg/dL are reported to have maximum urine osmolalities between 350 and 600 mOsm/L. Hypercalciuria in the absence of hypercalcemia also results in impaired renal concentrating ability. Interestingly, patients with familial hypocalciuric hypercalcemia do not have a renal concentration defect, suggesting that the hypercalciuria is the main contributing factor. The renal concentrating defect is reversible with correction of hypercalcemia and hypercalciuria.
Animal studies have revealed that hypercalcemia leads to a renal concentrating defect with polyuria and polydypsia via downregulation of AQP2 and AQP3. The mechanism behind this appears to be via the epithelial calcium-sensing receptor (CaSR). The CaSR is a G-protein that is expressed in renal epithelial cells that is activated by calcium. It has been demonstrated in a mouse cortical collecting duct cell line that luminal calcium activates the CaSR, resulting in reduced efficiency of coupling between the V2 receptor adenylate cyclase, leading to decreased cAMP levels and AQP2 expression. Basolateral calcium does not lead to a similar effect, explaining why hypercalciuria in the absence of hypercalcemia results in a concentrating defect in humans and why hypercalcemia in the absence of hypercalciuria does not. An additional mechanism, which may play a role in the polyuria associated with hypercalcemia, is the downregulation of sodium transporters presumably in response to activation of the CaSR.
Central diabetes insipidus (CDI) is defined as a urinary concentrating defect resulting from insufficient secretion of AVP in response to osmolar stimuli. CDI can be of varying severity depending on the etiology of disease. There are a number of defects that can affect the neurosecretory pathway for AVP secretion that can result in CDI. The cause of CDI can be broadly classified as either due to genetic mutations, where there is abnormal synthesis or packaging of AVP, or congenital or acquired damage to the AVP-producing magnocellular neurons or the supraopticohypophyseal tract. Approximately 50% of cases of CDI in children and young adults are idiopathic.
Hereditary forms of CDI are rare. The most common is that of autosomal dominant neurohypophyseal DI, which is caused by mutations in the neurophysin-vasopressin II (AVP-NPII) gene. Over 50 heterozygous mutations have been described. The symptoms of CDI are absent at birth but develop months to years later and are progressive throughout life. It is believed that these mutations affect amino acid residues that play a critical role in removing the signal residue and/or the proper folding of the prohormone in the endoplasmic reticulum of the neuron. The accumulation of these mutant prohormones is believed to be cytotoxic to the AVP-producing magnocellular neurons located in the supraoptic and paraventricular nuclei, leading to progressive cell damage. Autosomal recessive neurohypophyseal diabetes insipidus has a similar clinical presentation as the autosomal dominant form, being absent at birth and presenting symptoms later in life which are progressive. The mechanism of DI is different for the autosomal recessive form than the dominant form. In the recessive form, there is a mutation in the prohormone that does not interfere with folding or transportation of the prohormone and does not result in destruction of magnocellular neurons. Rather, the mutation results in a mutant AVP hormone that is secreted but has little or no antidiuretic effect. An X-linked recessive form of DI has been reported. It affects males and results in a progressive loss of AVP production over time. Wolfram syndrome, also known as DIDMOAD syndrome, is a rare autosomal recessive neurodegenerative condition that typically presents in childhood and is characterized by DI, diabetes mellitus, optic atrophy, and deafness. Families with Wolfram syndrome have mutations in a gene encoding the WFS1 protein, located in 4p16.1 regions, a transmembrane protein in the endoplasmic reticulum expressed in numerous tissues.
Congenital central nervous system (CNS) malformations are the cause of CDI in approximately 5% of unselected children with CDI. There are a variety of congenital malformations which can result in CDI ( Table 8.2 ), the most important of which involve midline defects and, in particular, septo-optic dysplasia. The pituitary gland develops at a very early stage of embryonogenesis and is closely linked to forebrain development. There is a strong association between developmental abnormalities of the pituitary and the forebrain. Septo-optic dysplasia is defined as the combination of optic nerve hypoplasia, midline neuroradiologic abnormalities, and pituitary hypoplasia with hypopituitarism. There is great phenotypic variability and the diagnosis is usually made with the presence of two of the three features. A variety of gene defects have been implicated in this condition. CDI is present in approximately 20% of the patients with septo-optic dysplasia.
Genetic |
|
Congenital |
|
Acquired |
|
Craniopharyngioma, germinoma, meningioma, optic glioma, pituitary adenoma, CNS lymphoma, leukemia, metastases |
|
Langerhans cell histiocytosis, neurosarcoidosis, tuberculosis, Wegener’s disease, xanthoma disseminatum |
|
Internal carotid aneurysm, hypoxic/ischemic brain injury, Sheehan’s syndrome, cerebral hemorrhage or infarction |
|
Chronic meningitis, viral encephalitis, congenital cytomegalovirus, or toxoplasmosis |
|
Lymphocytic infundibuloneurohypophysitis |
Autoantibodies to AVP-secreting cells |
|
Snake venom |
|
CDI is most commonly an acquired condition due to various causes. Thirty to fifty percent of the cases of CDI are idiopathic. The most common identifiable causes of CDI are brain tumors, granulomatous diseases, autoimmune or vascular diseases, traumatic brain injury (TBI), subarachnoid hemorrhage, or following brain surgery.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here