Pure water loss

Total body solute (TBS) is the product of TBW and plasma osmolality (P_osm). If a patient’s fluid loss has been limited to pure water, the following relationship is true:

If we assume that body water (TBW) is 60% of body weight measured in kilograms (Wt) and that 2.1 × P_Na is an estimate of P_osm:

This equation reduces to:

The water deficit is the difference between the previous and present body weights:

or

Consider a previously normal dog that has been deprived of water for several days. The dog weighs 10 kg at presentation, and its serum sodium concentration is 170 mEq/L. Assuming a previously normal serum sodium concentration of 145 mEq/L, the dog’s water deficit can be calculated:

The original estimates of TBW and serum sodium concentration may be modified based on information available to the clinician at presentation. For example, if the dog’s normal serum sodium concentration is known from a previous admission, this value can be substituted in place of 145 mEq/L. If the dog’s previous normal body weight is known, the water deficit may simply be estimated as the difference between the previous and present body weights. The assumption inherent in the latter calculation is that the patient has not gained or lost tissue mass. For a short period, this is a reasonable assumption because loss of 1 kg of tissue mass requires an expenditure of approximately 1600 kcal. This caloric expenditure would require fasting for 2 to 3 days in a normal 10-kg dog with a basal energy requirement of approximately 700 kcal.

A pure water deficit can be replaced by giving 5% dextrose in water intravenously. This solution technically is only slightly hypotonic to plasma (278 mOsm/kg), but the glucose ultimately enters cells and is metabolized so that administration of 5% dextrose is equivalent to administration of water. The water deficit must be replaced and hypernatremia corrected slowly over 48 hours. The brain adapts to hypertonicity by the production of osmolytes or idiogenic osmoles that prevent cellular dehydration. Excessively rapid lowering of the serum sodium concentration may result in movement of water into brain cells and development of cerebral edema. In human patients with hypernatremia of chronic or unknown duration, correction of the serum sodium concentration at a rate of less than 10 to 12 mEq/L per 24 hours minimizes the risk of neurologic complications related to water intoxication.^6,152 The animal’s serum sodium concentration should be monitored serially during replacement of the water deficit.

Hypotonic loss

As described earlier, hypotonic losses cause more severe extracellular volume contraction than do losses of pure water. As the tonicity of the fluid lost approaches the tonicity of ECF, the extracellular volume deficit becomes greater (see Fig. 3-9). As a result, signs of volume depletion are more likely with hypotonic losses, and the original replacement fluid should be isotonic so that extracellular volume repletion can proceed rapidly.

In the presence of hemorrhagic shock, whole blood, plasma, or a colloid solution is the ideal fluid to administer. The hemoglobin in whole blood improves oxygen-carrying capacity. The plasma proteins in whole blood and plasma or the dextrans in a colloid solution increase and maintain intravascular volume by increasing oncotic pressure. In many animals that have experienced severe hypotonic losses over an extended time period, replacement of the ECF volume with an isotonic crystalloid solution (e.g., 0.9% NaCl and lactated Ringer’s solution) is adequate. A volume up to four times the suspected intravascular deficit may be required because the isotonic crystalloid solution distributes rapidly throughout the ECF compartment (ECF volume is four times intravascular volume). After the extracellular volume has been expanded, hypotonic fluids (e.g., 0.45% NaCl and half-strength lactated Ringer’s solution) can be administered to provide fluids for maintenance needs and ongoing losses (see Chapter 14).

Gain of impermeant solute

The patient with an excess of sodium-containing impermeant solute in the ECF can be treated by administration of 5% dextrose intravenously. The main disadvantage of this approach is that it causes further expansion of the extracellular compartment in a patient already suffering from ECF volume expansion. In an animal with normal cardiac and renal function, this volume expansion leads to diuresis and natriuresis, and ECF volume returns to normal. In an animal with underlying cardiac disease or oliguria related to primary renal disease, this approach may lead to development of pulmonary edema. Administration of a loop diuretic (e.g., furosemide and ethacrynic acid) promotes excretion of the existing sodium load and hastens return of ECF volume to normal. As in the case of pure water deficit, it is essential that fluid administration proceeds slowly and that serum sodium concentration be lowered gradually over 48 hours to avoid neurologic complications.

Hyponatremia with normal plasma osmolality

Sodium is present as charged particles in the aqueous phase of body fluids. Approximately 93% of plasma volume is occupied by water, and the remaining 7% consists largely of proteins and lipids. Historically, serum sodium concentration has been measured by flame photometry. Flame photometry measures the number of sodium ions in a specific volume of plasma or serum. Thus, the sodium concentration is measured as if the sodium ions were present throughout the entire sample volume, whereas actually they are active only in the aqueous phase. Normally, this error is small. In plasma or serum samples containing a large amount of lipid or protein, however, the error may be larger, and the decrease in measured serum sodium concentration could be misleading to the clinician (Fig. 3-13). When serum sodium concentration is measured by direct potentiometry using ion-selective electrodes, large amounts of lipid or protein in the sample should not affect the measured serum sodium concentration. However, if the serum sample is diluted before measurement, large amounts of lipid or protein may still affect the measured serum sodium concentration.⁸⁹ Therefore, the clinician must be familiar with the laboratory method used so as to interpret serum sodium concentrations properly. The occurrence of a decreased serum sodium concentration as a result of laboratory methodology in the presence of normal plasma osmolality is called pseudohyponatremia or factitious hyponatremia. Pseudohyponatremia occurs in conditions associated with hyperlipidemia or severe hyperproteinemia.

Figure 3-13 Effect of increased plasma lipids on serum sodium concentration (pseudohyponatremia or factitious hyponatremia).

(From DiBartola SP. Hyponatremia. Vet Clin North Am Small Anim Pract 1998;28:515–532.)

Plasma osmolality in patients with pseudohyponatremia is normal, because lipids and proteins are very large molecules that contribute very little to plasma osmolality. If pseudohyponatremia is present, the calculated plasma osmolality is low because of a spuriously low serum sodium concentration, whereas the measured osmolality is normal. Thus, when an abnormal osmolal gap is present and the measured osmolality is normal, pseudohyponatremia should be suspected. The diagnosis of pseudohyponatremia can be made by visual inspection of plasma for lipemia and by measurement of the total plasma protein concentration. Hyperlipemia severe enough to cause pseudohyponatremia is visible to the naked eye as lactescent plasma. Each milligram per deciliter of lipid in serum reduces the sodium concentration by 0.002 mEq/L (e.g., a serum triglyceride concentration of 1000 mg/dL would be expected to reduce the serum sodium concentration by 2 mEq/L).¹¹⁸ In the case of hyperproteinemia, each gram per deciliter of protein above a concentration of 8 g/dL reduces the serum sodium concentration by approximately 0.25 mEq/L (e.g., the serum sodium concentration of a patient with a serum protein concentration of 12 g/dL would be expected to be reduced by 1 mEq/L).¹⁵¹ At such protein concentrations, the plasma may be viscous, and this is likely to occur mainly in patients with plasma cell dyscrasias. Thus, whereas pseudohyponatremia may be intellectually interesting, it is unlikely to be of clinical relevance in most instances. Furthermore, pseudohyponatremia itself has no consequences for the health of the patient. Its importance lies in the ability of the clinician to recognize it and refrain from treating the patient for hyponatremia. Treatment should be directed at the underlying disorder causing hyperproteinemia or hyperlipidemia.

Hyponatremia with increased plasma osmolality

If an impermeant solute is added to ECF, water moves from ICF to ECF, and the osmolality of both compartments increases (see Fig. 3-11).⁴¹ If the added solute is something other than sodium, the serum sodium concentration is reduced by the translocation of water, but the plasma osmolality is higher than normal.

Hyponatremia with hyperosmolality is usually caused by hyperglycemia in diabetes mellitus, wherein each 100-mg/dL increase in glucose may decrease the serum sodium concentration by 1.6 mEq/L.⁸² This correction factor worked well up to a blood glucose concentration of 440 mg/dL in a study in normal humans made transiently hyperglycemic by infusion of somatostatin, but the correction factor was much greater at higher blood glucose concentrations.⁷³ The authors concluded that an overall correction factor of a 2.4-mEq/L decrement in sodium for each 100-mg/dL increment in glucose would be preferable. In the diabetic patient, both hyperlipidemia and hyperglycemia may contribute to decreased serum sodium concentration. Administration of the osmotic diuretic mannitol also can cause hyponatremia with plasma hyperosmolality. The calculated osmolality is normal, the measured osmolality is high, and the osmolal gap is increased in the presence of mannitol, which is an unmeasured osmole. Hyperglycemia does not affect the osmolal gap because the plasma glucose concentration is part of the equation used to calculate plasma osmolality (i.e., it is a measured osmole).

Initially, TBW content is not altered in the setting of hyponatremia with hyperosmolality. Rather, there is an altered distribution of water between intracellular and extracellular compartments. However, a reduction in TBW content develops to the extent that these substances cause an osmotic diuresis.

Hyponatremia with decreased plasma osmolality

The total body sodium content and ECF volume of patients with hyponatremia and hypoosmolality may be normal, decreased, or increased, and hyponatremia may be classified according to the volume status of the patient as hypovolemic, normovolemic, and hypervolemic. In most instances, nonosmotic stimulation of antidiuretic hormone results in water retention and development of hyponatremia. Therefore, the second step in the evaluation of the patient with hyponatremia is to estimate total body sodium content and ECF volume status. This is best done by clinical assessment of the patient based on history, physical examination, and a few ancillary tests. A good history often indicates a source of fluid loss (e.g., vomiting, diarrhea, or diuretic administration), and the physical examination provides important clues to the patient’s volume status. The following physical findings should be assessed: skin turgor, moistness of the mucous membranes, capillary refill time, pulse rate and character, appearance of the jugular veins (distended or flat), and presence or absence of ascites or edema. Measurements of hematocrit and total plasma protein concentration, as well as systemic blood pressure and central venous pressure determinations, if available, further clarify the patient’s ECF volume status.

Hypovolemic Hyponatremia (Hyponatremia with Volume Depletion)

For a patient with volume depletion (hypovolemia) to develop hyponatremia, the total body deficit of sodium must exceed that of water. Hyponatremic patients with volume depletion have lost fluid by renal or nonrenal routes. Gastrointestinal losses (e.g., vomiting, diarrhea) and third-space losses, such as pleural effusion or peritoneal effusion caused by peritonitis, pancreatitis, or uroabdomen, are the most important nonrenal losses of fluid and NaCl.^19,162 Gastrointestinal losses are often hypotonic in nature. The question thus arises, “If the losses are hypotonic, how does the patient become hyponatremic?” The answer follows from three physiologic events and reflects the body’s tendency to preserve volume at the expense of tonicity. First, volume depletion decreases GFR, enhances isosmotic reabsorption of sodium and water in the proximal tubules, and decreases delivery of tubular fluid to distal diluting sites. These events impair excretion of water. Second, volume depletion is a strong nonosmotic stimulus for vasopressin release, and the increased plasma vasopressin concentration further impairs water excretion. Third, the patient is thirsty because of volume depletion and continues to drink water if it is available. All of these factors have a dilutional effect on the remaining body fluids. In one study, approximately 20% of dogs with gastrointestinal foreign bodies had hyponatremia.¹⁰

Recall the previous example of the loss of 1 L of fluid with an osmolality of 150 mOsm/kg and consider what would happen if the animal in question drinks 1 L of pure water after sustaining the hypotonic loss. The added water increases the ECF volume from 1.36 to 2.36 L, and the resulting hypotonicity rapidly drives water into cells to equalize osmolality:

where x is the volume of water moving between compartments:

The new volumes and osmolalities are:

Note that in this example the intracellular compartment is expanded (4.36 L). The volume of the extracellular compartment (1.64 L) is greater than it was when the same hypotonic loss was not replaced (1.36 L) but still less than the previous normal value (2 L). Thus, hypotonic (or isotonic) losses replaced by pure water lead to expansion of the ICF space. These changes are depicted in Figure 3-14.

Figure 3-14 Effect of drinking 1 L of water after a loss of 1 L of hypotonic fluid (150 mOsm/kg) on volume and tonicity of extracellular fluid (ECF) and intracellular fluid (ICF).

(Drawing by Tim Vojt.)

Renal fluid and NaCl losses resulting in hyponatremia are usually caused by hypoadrenocorticism or diuretic administration. In one study, 81% of 225 dogs with hypoadrenocorticism were hyponatremic at presentation.¹²⁹ Mineralocorticoid deficiency in hypoadrenocorticism results in urinary loss of NaCl and depletion of ECF volume. Volume depletion in patients with hypoadrenocorticism is a strong nonosmotic stimulus for vasopressin release and impairs water excretion.⁶ Hyperkalemia typically accompanies hyponatremia in hypoadrenocorticism.^{129,132,144,161} However, some dogs with hypoadrenocorticism have only glucocorticoid deficiency at the time of presentation and thus have normal serum potassium concentrations.^100,139,156 Glucocorticoids are necessary for complete suppression of vasopressin release, and in their absence impaired water excretion and hyponatremia can occur.²⁸ Occasionally, dogs with gastrointestinal fluid losses develop electrolyte disturbances that mimic hypoadrenocorticism.^30,107 Hyponatremia associated with third-space loss of fluid has been reported with pleural effusion related to chylothorax, lung lobe torsion, and neoplasia.^{93,155,162,166} In these reports, hyponatremia was attributed at least in part to removal of sodium-rich fluid by thoracocentesis. However, many of these animals had evidence of volume depletion, and it is likely that nonosmotic vasopressin secretion also played a role in the development of hyponatremia. Affected dogs also had mild hyperkalemia attributed to decreased renal excretion of potassium caused by volume depletion and decreased distal renal tubular flow. Similar findings have been observed in dogs and cats with peritoneal effusion and in dogs in late pregnancy.^9,85,145 The pathogenesis of hyponatremia and mild hyperkalemia in dogs with gastrointestinal losses is probably similar to that described for dogs with pleural and peritoneal effusions, but the explanation for the rare occurrence of similar electrolyte abnormalities in dogs in late pregnancy is unknown. When the cause of hyponatremia and hyperkalemia is unclear, an ACTH stimulation test should be performed to rule out hypoadrenocorticism.

Diuretics contribute to impaired water excretion and dilution of sodium in the ECF by decreased distal delivery of tubular fluid and nonosmotic stimulation of vasopressin release, which occur in response to volume depletion. Furthermore, potassium depletion caused by diuretics can contribute to hyponatremia because shifting of intracellular potassium into the extracellular compartment in exchange for sodium may occur. Hyponatremia has been associated with chronic blood loss in dogs.¹⁵⁸ It was thought that defective urinary concentrating ability in these dogs was caused by impaired vasopressin release in response to plasma hypoosmolality and loss of NaCl from the renal medullary interstitium. Some of these dogs had hypoadrenocorticism and gastrointestinal fluid losses that might have contributed to their hyponatremia. Normal concentrating ability returned after resolution of hyponatremia.

Treatment of hyponatremia

The two main goals of treatment in hyponatremia are to diagnose and manage the underlying disease and, if necessary, increase serum sodium concentration and plasma osmolality. Severe, symptomatic hyponatremia of acute onset (<24 to 48 hours’ duration) may result in seizures, cerebral edema, or death and requires prompt treatment. In human patients with acute hyponatremia, correction of serum sodium concentration may be required at rates up to 12 mEq/L/day.¹⁵¹ However, severe, symptomatic hyponatremia of rapid onset is rare in small animal practice. Because of inexperience with the management of acute hyponatremia in dogs and cats, and the known risks of overly rapid correction of hyponatremia, only use of conventional crystalloid solutions (e.g., lactated Ringer’s solution and 0.9% saline) is recommended. Use of 3% NaCl is not recommended.

Patients with chronic hyponatremia often have few or no clinical signs directly attributable to their hypoosmolality. This is probably because the brain has had sufficient time to adapt to plasma hypotonicity. In fact, treatment of chronic hyponatremia can be more dangerous than the disorder itself. In human patients, complications of treatment may occur when chronic (>48 hours’ duration) hyponatremia is corrected too rapidly (i.e., when the serum sodium concentration is increased by >10 to 12 mEq/L in 24 hours).^95,151

When hyponatremia and hypoosmolality are corrected, potassium and organic osmolytes lost during adaptation must be restored to the cells of the brain. If replacement of these solutes does not keep pace with the increase in serum sodium concentration that occurs as a result of treatment, brain dehydration and injury—called osmotic demyelination or myelinolysis—may result.^95,151 Experimental studies have confirmed that this syndrome is a result of a rapid and large increase in serum sodium concentration and is not a consequence of hyponatremia and hypoosmolality. Human patients with hyponatremia of more than 72 hours’ duration are more susceptible than those with hyponatremia of less than 24 hours’ duration.¹⁵¹ The neural lesions of myelinolysis develop several days after correction of hyponatremia and consist of myelin loss and injury to oligodendroglial cells in the pons and other sites in the brain (e.g., thalamus, subcortical white matter, and cerebellum). Lesions may take several days to develop, but on magnetic resonance imaging they are hyperintense on T2-weighted images, hypointense on T1-weighted images, and are not enhanced after gadolinium injection.⁹⁵ The ability to reaccumulate organic osmolytes may vary among different regions of the brain and thus account for why some regions (e.g., midbrain) are more susceptible to osmotic demyelination.

Similar lesions have been reported in experimental dogs with hyponatremia with correction rates of 15 mEq/L/day even without overcorrection to hypernatremia.⁹⁴ In veterinary medicine, myelinolysis first was reported in two dogs after correction of hyponatremia associated with trichuriasis.¹²² In one dog, a serum sodium concentration of 101 mEq/L had been corrected to 136 mEq/L in less than 38 hours (correction rate, >22 mEq/L/day), and in the other, a serum sodium concentration of 108 mEq/L had been corrected to 134 mEq/L in less than 38 hours (correction rate, >16 mEq/L/day). Clinical signs developed 3 to 4 days after correction of hyponatremia and consisted of lethargy, weakness, and ataxia progressing to hypermetria and quadriparesis. Lesions were detected by magnetic resonance imaging and were located in the thalamus as compared with the more typical pontine location in affected human patients. From this experience, it was recommended that dogs with asymptomatic chronic hyponatremia be treated by mild water restriction and monitoring of serum sodium concentration. Symptomatic dogs with chronic hyponatremia should be treated conservatively at correction rates of less than 10 to 12 mEq/L/day (0.5 mEq/L/hr). Serial monitoring of serum sodium concentration is necessary because the actual rate of correction may not correspond to the calculated rate of correction. Correction should be carried out with conventional crystalloid solutions (e.g., lactated Ringer’s solution and 0.9% NaCl) in a volume calculated specifically to replace the patient’s volume deficit. The clinician must remember that volume repletion in hypovolemic patients abolishes the nonosmotic stimulus for vasopressin release and allows the animal to excrete solute-free water via the kidneys. This in itself tends to correct the hyponatremia. Thus, caution should be exercised even when using conventional crystalloid fluid therapy.

Three additional cases of suspected myelinolysis in dogs with chronic hyponatremia caused by hypoadrenocorticism or trichuriasis have been reported.^11,24,105 The rates of correction of hyponatremia in these dogs were 22 mEq/L on day 1 and 17 mEq/L on day 2, 32 mEq/L over 2 days, and 17 mEq/L in 9 hours.^11,24,105 The neurologic signs that developed (e.g., spastic tetraparesis, loss of postural and proprioceptive responses, dysphagia, trismus, and decreased menace response) were similar to those originally described by O’Brien.¹²² The dogs of these reports gradually recovered over several weeks.

Water intake should be carefully restricted to a volume less than urine output in normovolemic patients with hyponatremia (e.g., psychogenic polydipsia), or drugs causing an antidiuretic effect should be discontinued if possible. Demeclocycline and lithium inhibit vasopressin release and have been used to treat SIADH in humans, but water restriction is probably the safest approach.⁴⁸

In edematous patients, dietary sodium restriction and diuretic therapy should be considered. A 0.9% NaCl solution can be administered concurrently with loop diuretics (e.g., furosemide) to effect more rapid correction of hyponatremia in overhydrated symptomatic patients. The occurrence of chronic hyponatremia in patients with congestive heart failure is often a sign of advanced disease and responds poorly to treatment. Administration of furosemide and an angiotensin-converting enzyme inhibitor (e.g., enalapril) may improve stroke volume and cardiac output by reducing preload and afterload and may decrease vasopressin secretion and enhance water excretion, which in turn may facilitate resolution of hyponatremia.

Arginine vasopressin (AVP) receptor antagonists (vaptans) block either V₂ receptors (lixivaptan, tolvaptan, satavaptan) or both V₂ and V_1A receptors (conivaptan).⁎ Based on their mechanism of action, these drugs increase free water excretion by the kidneys and effectively normalize serum sodium concentration in patients with non-osmotic release of AVP causing euvolemic (e.g., SIADH) or hypervolemic (e.g., congestive heart failure, liver failure) hyponatremia.^111,117 Patients with hypovolemic hyponatremia should be treated with an infusion of 0.9% NaCl or other isotonic fluid to replace their volume deficits. The AVP receptor antagonists increase water but not solute excretion via the kidneys, and likely will have major impact on the clinical management of euvolemic and hypervolemic hyponatremia in the near future.⁶⁰

Conivaptan is administered as an intravenous bolus in 5% dextrose followed by a constant rate infusion, whereas tolvaptan, lixivaptan, and satavaptan are administered orally. In humans, adverse effects of the vaptans generally are limited to thirst and dry mouth. To minimize the risk of osmotic demyelination, correction of serum sodium concentration should be limited to <8 mEq/L per 24 hours, especially when hyponatremia is chronic or of unknown duration. The pH of the conivaptan solution is low (3.0) and infusion site reactions are common.

Because they antagonize only the V₂ receptor, lixivaptan and tolvaptan do not cause changes in blood pressure. Vasopressin receptor antagonists promote aquaresis and correct hyponatremia in heart failure patients, but long-term beneficial effects on patient survival have not yet been documented.^21,40 Conivaptan may be especially helpful in heart failure patients as a result of its effects of the V_1A receptor and potential to decrease total peripheral resistance.⁵⁴ Although the combined V₂/V_1A receptor antagonist conivaptan might be expected to lower blood pressure, it only causes hypotension in 2.5% of treated patients, although orthostatic hypotension may be seen in 5% of treated patients.¹²⁴

Conivaptan is both a substrate for and inhibitor of CYP3A4 (the 3A4 isoform of the cytochrome P-450 enzyme). Administration of conivaptan with other CYP3A4 inhibitors (e.g., ketoconazole, itraconazole, clarithromycin) will markedly increase its plasma concentration, and use with other drugs metabolized by CYP3A4 (e.g., statins, midazolam, amlodipine) will increase the plasma concentrations of these drugs and potentially lead to toxicity.

Although considered primarily aquaretics, tolvaptan and lixivaptan may increase sodium excretion at higher dosages, possibly by blocking the sodium-retaining effect of AVP on the thick ascending limb of Henle’s loop. Unresolved is whether or not V₂-receptor antagonists could decrease concentrations of factor VIII and von Willebrand factor, which are known to be increased by AVP.

Endogenous creatinine clearance

In chronic progressive renal disease, urinary concentrating ability is impaired after two thirds of the nephron population has become nonfunctional, whereas azotemia does not develop until three quarters of the nephrons have become nonfunctional. Thus, the main indication for determination of endogenous creatinine clearance is the clinical suspicion of renal disease in a patient with polyuria and polydipsia but normal BUN and serum creatinine concentrations. The only requirements for determination of endogenous creatinine clearance are an accurately timed collection of urine (usually 24 hours), determination of the patient’s body weight, and measurement of serum and urine creatinine concentrations. Failure to collect all urine produced results in an erroneously reduced calculated clearance value. Use of creatinine clearance as an estimate of GFR is discussed further in Chapter 2.

Water deprivation test

The water deprivation test is indicated in evaluation of animals with confirmed polydipsia and polyuria, the cause of which remains undetermined after the initial diagnostic evaluation. It is usually performed in animals with hyposthenuria (USG <1.007) that are suspected to have CDI, NDI, or psychogenic polydipsia. An animal that is dehydrated but has dilute urine has already failed the test and should not be subjected to water deprivation. In such an animal, failure to concentrate urine is probably caused by structural or functional renal dysfunction or administration of drugs that interfere with urinary concentrating ability. The water deprivation test is also contraindicated in animals that are azotemic. The test should be performed with extreme caution in animals with severe polyuria because such patients may rapidly become dehydrated during water deprivation if they have defective urinary concentrating ability.

At the beginning of the water deprivation test, the bladder must be emptied and baseline data collected (body weight, hematocrit, total plasma proteins, skin turgor, serum osmolality, urine osmolality, and USG). Water is then withheld, and these parameters are monitored every 2 to 4 hours. Urine and serum osmolalities are the best parameters to follow, but osmolality results are often not immediately available to the clinician. Thus, USG and body weight assume great importance for decision making during performance of the test. An increase in total plasma protein concentration is a relatively reliable indicator of progressive dehydration, but increases in hematocrit and changes in skin turgor are less reliable.⁶⁷ Serum creatinine and BUN concentrations should not increase during a properly conducted water deprivation test.

The bladder should be emptied at the time of each urine collection. Maximal stimulation of ADH release is present after loss of 5% of body weight. The test is concluded when the patient either demonstrates adequate concentrating ability or becomes dehydrated as evidenced by loss of 5% or more of its original body weight. It is important when weighing the animal to use the same scale each time and to empty the bladder at each evaluation.

In normal dogs, dehydration becomes evident after a mean of 42 hours but occasionally may not occur until after 96 hours.⁶⁷ The time required for dehydration to develop during water deprivation testing in dogs with disorders characterized by polyuria and polydipsia may be as short as a few hours or up to 12 hours. By the time dehydration is evident, normal dogs develop a USG of 1.050 to 1.076, urine osmolality of 1787 to 2791 mOsm/kg, and a urine/plasma osmolality ratio of 5.7 to 8.9.⁶⁷ Normal cats developed USG values of 1.047 to 1.087 and urine osmolalities of 1581 to 2984 mOsm/kg after water deprivation of sufficient duration (approximately 40 hours) to induce 5% loss of body weight.¹⁴² Failure to achieve maximal urinary solute concentration does not localize the level of the malfunction, but a structural or functional defect may be present anywhere along the hypothalamic-pituitary-renal axis. Furthermore, animals with renal medullary solute washout may have impaired concentrating capacity regardless of the underlying cause of polyuria and polydipsia.

If there has been less than a 5% increase in urine osmolality or less than 10% change in USG for three consecutive determinations or if the animal has lost 5% or more of its original weight, 0.25 to 0.5 U/kg aqueous vasopressin (pitressin) (up to a total dose of 5 U) or 5 µg of DDAVP may be given subcutaneously and parameters of urinary concentrating ability monitored at 30, 60, and 120 minutes after ADH injection. Normal dogs and those with psychogenic polydipsia should show no additional response to ADH administration in this setting. The expected responses to water deprivation for dogs with various disorders of water balance are shown in Figure 3-16.

Figure 3-16 Idealized response to water deprivation of dogs with various disorders of water balance. CDI, Central diabetes insipidus; NDI, nephrogenic diabetes insipidus; MSW, medullary solute washout; ADH, antidiuretic hormone.

Modified water deprivation test

A modified water deprivation test has been described for the diagnosis of polyuric disorders in dogs. Water is removed from the animal’s cage and the urinary bladder emptied, after which urine osmolality or specific gravity is measured and the bladder emptied on an hourly basis. Maximal urine solute concentration is defined as occurring whenever less than a 5% increase in urine osmolality occurs on sequential determinations. This maximal concentration occurred at a mean urine osmolality of 1414 mOsm/kg in normal dogs after 24 hours of water deprivation.¹¹⁶ At this time, 2 to 3 U of aqueous vasopressin was administered subcutaneously and the urine osmolality determined at 1 and 2 hours after injection. Further increase in urine osmolality after administration of vasopressin should not exceed 10% in normal dogs. In this study, dogs with CDI showed an average 292% increase in urine osmolality after aqueous ADH, dogs with partial CDI an average 28% increase, and dogs with hyperadrenocorticism an average 20% increase. The time required to develop dehydration ranged from 3 to 11.5 hours in dogs with psychogenic polydipsia, complete or partial CDI, and hyperadrenocorticism.

Gradual water deprivation

Gradual water deprivation can be performed to eliminate diagnostic confusion caused by renal medullary solute washout. The owner can be instructed to restrict water consumption to 120 mL/kg/day 72 hours before, to 90 mL/kg/day 48 hours before, and to 60 mL/kg/day 24 hours before the scheduled water deprivation test. In dogs with psychogenic polydipsia, this promotes release of endogenous vasopressin, increased permeability of the inner medullary collecting ducts to urea, and restoration of the normal gradient of medullary hypertonicity. An alternative approach is to instruct the owner to reduce water consumption by approximately 10% per day over a 3- to 5-day period (but not to less than 60 mL/kg/day). This approach should be used only in animals that are otherwise healthy on initial clinical evaluation, and the owner should provide dry food ad libitum and weigh the dog daily to monitor for loss of body weight.

Hickey-hare test

In the Hickey-Hare test, water (20 mL/kg) is administered by stomach tube, an indwelling urinary catheter is placed, and urine flow (milliliters per minute) is determined.⁹² Hypertonic saline (2.5%) is administered intravenously at a rate of 0.25 mL/min/kg for 45 minutes. Urine volume is recorded every 15 minutes during the infusion and for 45 minutes afterward. The normal response to this procedure is a decrease in the rate of urine production caused by stimulation of ADH release by plasma hyperosmolality. It is useful in the differentiation of psychogenic polydipsia with renal medullary solute washout from NDI after negative water deprivation and exogenous ADH test results. In NDI, there should be no change or an actual increase in urine flow, whereas in psychogenic polydipsia with renal medullary solute washout, repletion of solute (e.g., NaCl) should have occurred, and the response to hypertonic saline should be normal (decreased urine volume). This test is cumbersome, is contraindicated for patients with congestive heart failure, and may lead to signs of hypernatremia in patients that cannot excrete a sodium load. It has largely been replaced by gradual water deprivation as described previously.

Exogenous antidiuretic hormone testing

The exogenous vasopressin test may be used for debilitated patients in which water deprivation is considered hazardous or to further characterize a concentrating defect detected by the routine water deprivation test. In the aqueous vasopressin test, an intravenous infusion of aqueous vasopressin (pitressin) at 10 mU/kg is given over 60 minutes. The bladder is emptied at the start of the study, and parameters of urinary concentrating ability are measured before and at 30-minute intervals for 3 hours after beginning the infusion. The bladder is emptied at each measurement. In one report, maximal response to aqueous vasopressin in water-loaded dogs usually occurred at 60 minutes (range, 30 to 90 minutes) and consisted of USG values of 1.012 to 1.033, urine osmolalities of 429 to 1437 mOsm/kg, and urine/plasma osmolality ratios of 1.5 to 5.1.⁶⁸ Water should be provided ad libitum during testing, but water loading should not be performed in clinical patients.

In the repositol vasopressin test, 3 to 5 U of vasopressin tannate in oil (pitressin tannate) is given intramuscularly, and the bladder is emptied 3 to 6 hours after injection. Parameters of urinary concentrating ability are measured before and at 6, 9, 12, and 24 hours after injection. Oral water loading must be avoided because of the danger of potentially lethal water intoxication.⁶⁹ Maximal response to repositol vasopressin occurred 8 to 12 hours after injection and consisted of USG values of 1.028 to 1.057, urine osmolalities of 1052 to 1850 mOsm/kg, and urine/plasma osmolality ratios of 3.9 to 6.7.⁶⁹

The standard or modified water deprivation test is the preferred initial test of urinary concentrating ability because mean maximal values are usually higher with this test and results are easier to interpret (Table 3-5). Why higher values for parameters of urinary concentrating ability are achieved with this test as compared with the exogenous vasopressin tests is unknown. Possible explanations include the actions of antidiuretic substances other than ADH that may be present in hydropenic individuals, the effect of slower renal medullary blood flow in dehydrated patients, and intensification of the medullary interstitial gradient in dehydrated individuals.

Table 3-5 Mean Maximal Values for Parameters of Total Urine Solute Concentration⁎