Syndrome of Inappropriate Antidiuretic Hormone Secretion in a Cat

Introduction

The syndrome of inappropriate antidiuretic hormone secretion (SIADH) is a condition in which the body secretes arginine vasopressin (AVP) in the absence of appropriate osmotic or volemic stimulus. An increase in serum osmolality or a decrease in plasma volume (as sensed by osmoreceptors in the brain or baroreceptors in the heart and blood vessels) normally stimulates the release of AVP from the posterior pituitary. This promotes water reabsorption from the distal tubules in the kidneys.^1,2 When AVP is secreted inappropriately, water is reabsorbed despite a low serum osmolality. This excessive accumulation of free water leads to hyponatremia.^1,3

Though SIADH is a common cause of hyponatremia in humans, it has rarely been reported in the veterinary literature; only six cases have been described in dogs.^4–11 This report describes SIADH in a cat, which, to our knowledge, has not previously been documented.

Case Report

A 3-year-old, 3.95-kg, spayed female domestic shorthaired cat was referred for evaluation of weight loss, inappetence, and elevated liver enzyme activities of 2 weeks’ duration. On presentation, the cat was quiet, alert, and febrile (39.7°C). Physical examination abnormalities included a body condition score of 4/9 and mild hepatomegaly. A normal complete blood count, elevated alkaline phosphatase activity, alanine aminotransferase activity, total bilirubin concentration, and low glucose and blood urea nitrogen (BUN) concentrations [Table 1] were present on initial testing. Prothrombin time (PT; 31.5 seconds, reference range 10.3 to 12.8 seconds) and partial thromboplastin time (PTT; >60 seconds, reference range 19 to 25 seconds) were markedly prolonged. Tests for feline leukemia virus and feline immunodeficiency virus were negative. Abdominal ultrasound revealed that the liver was mildly enlarged and hyperechoic.

Intravenous (IV) fluid therapy using 0.9% sodium chloride^a with 2.5% dextrose^b was administered at 10 mL per hour, and treatment with ursodeoxycholic acid^c (10 mg/kg per os q 24 hours) was initiated. Vitamin K₁d (1 mg/kg subcutaneously [SC] q 24 hours) and a transfusion of fresh-frozen plasma (FFP; 5 mL/kg IV) were given to treat the coagulopathy.

On day 2, the cat’s body weight had increased to 4.1 kg. Because of a persistently prolonged PT and PTT (PT 15.3 seconds, PTT >60 seconds), a second FFP transfusion (10 mL/kg IV) was administered. Following this treatment, the clotting times were within their reference ranges (PT 13.7 seconds, PTT 19.2 seconds). The cat was anesthetized using propofol^e (10 mg/kg IV) and was maintained with isoflurane^f in 100% oxygen. Using ultrasound guidance, bile was aspirated from the gallbladder for aerobic and anaerobic cultures. Multiple liver biopsies were obtained using a laparoscopic approach. Mucosal biopsies of the stomach and duodenum were obtained using endoscopy, and a percutaneous endoscopic gastrostomy (PEG) tube^g was placed. Treatment with amoxicillin and clavulanate potassium^h (15.6 mg/kg by tube q 12 hours) and S-adenosylmethionineⁱ (22.5 mg/kg by tube q 24 hours) was started pending biopsy and culture results. Intravenous fluid therapy with 0.9% sodium chloride and 2.5% dextrose was continued at 10 mL per hour. Liver and bile cultures were negative. Histopathology revealed hepatic lipidosis without an identifiable underlying cause and no remarkable abnormalities in the gastrointestinal biopsies.

On day 3 of hospitalization, the cat was quiet but alert, and she appeared icteric. Tube feedings were started using a 2 kcal/mL slurry.^j,k Eighteen mL of this slurry and 35 mL of water were given in divided feedings throughout the day. Caloric intake was gradually increased over the next 6 days, and the cat received between 24 and 117 mL per day (median 58 mL) of food and 20 to 45 mL per day (median 35 mL) of water. As feedings had been started, dextrose supplementation was discontinued, and fluid therapy was changed to Plasmalyte-148^l with vitamin B complex^m (2 mL/L) at 7 mL per hour. Because of persistent residual food volume in the stomach prior to feedings, metoclopramideⁿ (0.25 mg/kg SC q 6 hours) was added to the treatment regimen.

On the evening of day 4, the cat appeared depressed and lethargic. The blood glucose concentration was 273 mg/dL (reference range 80 to 155 mg/dL), and the systemic arterial blood pressure was 136/73 mm Hg (reference range 139±27/77±25).¹² By the following morning, the cat was extremely weak and ataxic with ventroflexion of the neck. Severe electrolyte imbalances (including hyponatremia, hypokalemia, and hypochloridemia), as well as other abnormalities associated with the cat’s liver disease, were found on plasma biochemical analyses [Tables 1, 2]. The urine specific gravity was 1.015, with a pH of 7.0 [Table 2]. Fluid therapy was changed to 0.9% sodium chloride supplemented with 40 mEq/L of potassium chloride^o and 2 mL/L of vitamin B complex at 11 mL per hour. Lactulose (0.5 mL/kg by tube q 8 hours) was started for possible hepatic encephalopathy.

On day 6, sodium, potassium, and chloride concentrations had all increased, although the sodium and chloride concentrations remained below the reference range. Potassium chloride supplementation in the fluids was decreased to 20 mEq/L. On day 7, sodium, potassium, and chloride concentrations dropped again and continued to decrease on day 8, despite increasing the potassium chloride supplementation to 40 mEq/L and increasing the fluid rate to 15 mL per hour. The cat experienced diarrhea after starting the lactulose, so this was discontinued because of lack of clinical improvement and concern it was contributing to the electrolyte disturbances. A plasma magnesium concentration was evaluated because of the refractory hypokalemia. Mild hypomagnesemia was found, and supplementation with magnesium sulfate^p (0.01 mEq/mg per hour IV) was instituted for 48 hours until the magnesium concentration was within the reference range. Silybin/vitamin E^q (2.25 mg/kg Silybin and 12.5 IU/kg vitamin E by tube q 24 hours) and L-carnitine^r (125 mg/kg by tube q 24 hours) were added as additional treatments for hepatic lipidosis. Vomiting was noted approximately once a day.

On day 9, hyponatremia, hypokalemia, and hypochloridemia persisted. The cat was weak, and mentation had become increasingly dull. Daily body weight measurements had not fluctuated by >50 g in the previous 8 days. The SIADH was suspected, and serum and urine osmolality and urine concentrations of sodium, potassium, and creatinine were measured from a single urine sample. The fractional excretions of sodium and potassium were calculated using previously described methods.^13,14 The serum was hyposmotic (272.3 mOsm/kg, reference range 290 to 330 mOsm/kg) with an inappropriately high urine osmolality of 412 mOsm/kg.¹ The urine sodium concentration was 133 mEq/L with a fractional excretion of 4.2%, and the urine potassium concentration was 40.2 mEq/L with a fractional excretion of 59%, making the fractional excretion of both electrolytes greater than previously reported values in normal cats.¹⁴ A potential limitation in using an isolated urine sample to evaluate the fractional excretion of electrolytes is that fractional excretion shows only moderate correlation to 24-hour urinary excretion of sodium and potassium in normal cats.¹⁴ Furthermore, IV infusions of sodium and potassium will cause increased excretion of these electrolytes.¹⁵ However, in the face of such severe electrolyte deficits, the magnitude of the renal losses was still considered inappropriate.

The clinical and laboratory findings were consistent with criteria for diagnosis of SIADH [Table 3]. Treatment with fluid restriction was started by decreasing the IV fluid rate to 8 mL per hour and continuing to decrease the rate over the next 5 days until it was discontinued on day 14. The amount of water used to flush the PEG tube was limited to 15 to 20 mL per day. The cat continued to receive 75 to 100 mL of food per day through the PEG tube. Metoclopramide was discontinued on day 9 because of concern it was contributing to inappropriate AVP secretion.¹⁶

Over the next 48 hours, the sodium and chloride concentrations increased, with normokalemia achieved on day 11. A venous blood gas analysis on day 11 identified a metabolic acidosis (pH 7.15, reference range 7.324 to 7.459; bicarbonate [HCO₃] 13.1 mmol/L, reference range 16.6 to 22.8 mmol/L; partial pressure of carbon dioxide [PCO₂] 39.1 mm Hg, reference range 27.7 to 41.2 mm Hg). The acidosis partially corrected (pH 7.232, HCO₃ 13.2 mmol/L, PCO₂ 32.6 mm Hg), but the hypokalemia worsened after administration of sodium bicarbonate^s (4 mEq IV over 8 hours). Potassium chloride supplementation was increased to 100 mEq/L in 0.9% sodium chloride at 4 mL per hour on days 11 through 13.

The cat was brighter by day 12, and mentation and strength gradually improved over the next week. Potassium gluconate^t (0.63 mEq/kg potassium by tube q 12 hours) was administered beginning day 13. From day 9 to 13, the cat lost 190 grams of body weight, presumably in free water. Following the discontinuation of IV fluids on day 14, water intake was restricted to 30 to 50 mL per day, and the cat received 85 to 120 mL of food per day through the PEG tube (approximately 40 mL/kg per day of total fluid intake). During this time, all electrolytes continued to increase. By day 15, the potassium and total carbon dioxide concentrations were within the reference ranges, and the cat remained only mildly hyponatremic and hypochloridemic. Oral potassium supplementation was tapered and stopped on day 21. At that time, water intake was increased to 120 mL per day, in addition to 120 mL of food per day (approximately 60 mL/kg per day of total fluid intake). The cat continued to show signs of ptyalism and occasional vomiting, especially when handled, so treatment with ranitidine^u (1 mg/kg by tube q 8 hours) and maropitant^v (0.5 mg/kg SC q 24 hours) was instituted. At the time of discharge on day 23, all electrolytes were within their respective reference ranges.

The cat was treated at home with amoxicillin and clavulanate potassium, ursodeoxycholic acid, SAM-e, Silybin/vitamin E, L-carnitine, ranitidine, tube feedings that provided 120 mL of food slurry and water each day, along with free-choice water and feeding. Follow-up 1 week later by the referring veterinarian revealed the electrolytes were still within the reference range, and most of the serum biochemical abnormalities had resolved. One month after the cat’s release from the hospital, all liver enzyme activities were within reference ranges. After 3 months, the PEG tube was removed, because the cat was eating and drinking normally without any signs of recurrent problems.

Discussion

Hyponatremia occurs when the amount of free water increases relative to sodium in the extracellular fluid. Hyponatremic animals can have increased plasma osmolality secondary to an increase in impermeable solutes (as seen with hyperglycemia or mannitol infusions) or normal plasma osmolality in cases of pseudohyponatremia (where sodium measures artifactually low because of laboratory methodology). However, most cases of hyponatremia are associated with plasma hyposmolality.^1,3

Causes of hyposmotic hyponatremia can be further categorized by extracellular volume into hyponatremia with volume depletion, volume excess, or normovolemia. Hypovolemic hyponatremia results from renal, gastrointestinal, or third space losses of fluid that lead to a decreased glomerular filtration rate, causing increased reabsorption of sodium and water in the proximal tubules but decreased delivery of fluid to the distal tubules, where excess water would normally be excreted. The hypovolemia causes baroreceptor-mediated stimulation of AVP, which promotes the reabsorption of free water.^1,3 Hypervolemic hyponatremia is a consequence of a decreased effective circulating volume, which is seen in cases of congestive heart failure, cirrhosis, and nephrotic syndrome. Similar to hypovolemic hyponatremia, this perceived hypovolemia causes decreased water excretion by the kidneys, stimulates secretion of antidiuretic hormone (ADH), and activates the reninangiotensin-aldosterone system, causing sodium and water retention.¹ Normovolemic hyponatremia can occur iatrogenically after the administration of hypotonic fluids or antidiuretic drugs, or it may occur as a result of psychogenic polydipsia, hypothyroid myxedema coma, or SIADH.^1,3

A diagnosis of SIADH is dependent upon meeting a series of criteria adopted from the human literature that includes: 1) hyponatremia with plasma hyposmolality; 2) inappropriately high urine osmolality (>100 mOsm/kg) in the face of plasma hyposmolality; 3) continued renal excretion of sodium (urine sodium >40 mEq/L); 4) normovolemia; 5) normal renal, adrenal, and thyroid function; 6) no history of diuretic administration; and 7) correction of hyponatremia by fluid restriction.^1,4–6 Measurement of plasma sodium concentrations, plasma and urine osmolality, and fractional excretion of sodium in this case fulfilled the first three of these prerequisites. No diuretics were administered, and resolution of hyponatremia with fluid restriction further supported the diagnosis.

Ideally, the cat’s volume status would have been evaluated by measuring central venous pressure, but, unfortunately, attempts to place a central line were unsuccessful. The cat appeared clinically well hydrated, had a normal systolic blood pressure, and maintained a stable body weight with adequate fluid supplementation; this suggested hypovolemia was not present. In addition, a hypovolemic state should have resulted in a low (<0.5%) fractional excretion of sodium.

Severe liver disease, as seen in the cat of this report, can cause hyponatremia because of the decreased effective circulating volume leading to nonosmotic release of ADH and impaired excretion of free water.¹ This is best described in humans and dogs with cirrhosis and concurrent ascites.^17,18,19 Liver biopsies of this cat did not show changes consistent with chronic, fibrotic lesions that would lead to this sequela. Edema and ascites were not appreciated on physical examination, nor did the cat gain weight as we might expect with fluid retention. Furthermore, unlike in this cat, animals that are hyponatremic from severe liver disease should have a low fractional excretion of sodium (<0.5%) because of aldosterone-induced sodium reabsorption.²⁰

No evidence of renal, adrenal, or thyroid disease was seen in this cat. The persistently low BUN and creatinine make kidney disease an unlikely cause of the observed electrolyte disturbances. Furthermore, a low BUN is a characteristic finding in humans with SIADH because of the increased fractional excretion of urea.²¹ Tests of adrenal and thyroid function were not performed, because clinical findings did not support any dysfunction. Both hypoadrenocorticism and hypothyroidism are rare in cats, and since the clinical signs resolved without treatment for either condition, it is unlikely that either endocrinopathy was present. Plasma AVP concentrations are generally of little utility in establishing a diagnosis of SIADH, as conditions that cause hypovolemia or a decreased effective circulating volume (e.g., heart failure, cirrhosis, nephrotic syndrome) will also cause increased AVP secretion from the activation of baroreceptors.⁶ Thus, the majority of animals with hyposmolar hyponatremia for any reason will have high levels of circulating AVP.⁶

Arginine vasopressin is secreted primarily in response to increased plasma osmolality, although hypovolemia can also act as a stimulus.^1,5,22 The AVP binds to vasopressin V2 receptors in the principal cells of the renal collecting duct and stimulates water reabsorption by promoting the insertion of aquaporin 2 water channels into the apical cell membranes.^1,5 When SIADH occurs, increased AVP leads to water reabsorption in the distal tubules that fails to respond to negative feedback from osmoreceptors and baroreceptors.⁵ The increase in extracellular fluid volume increases the arterial blood pressure, leading to pressure natriuresis and decreased reabsorption of sodium in the proximal tubules. This causes ongoing loss of sodium in the urine, as noted in the cat of this report.^23–25

Syndrome of inappropriate antidiuretic hormone secretion is rarely reported in dogs, and to our knowledge has not previously been reported in cats; however, SIADH is a very common cause of hyponatremia in humans. Pathological causes of SIADH in humans include malignancy, pulmonary disease, and disorders of the central nervous system.^4–6,24 Certain drugs (including diuretics, morphine, nonsteroidal antiinflammatory drugs, and 3,4-methylene-dioxymethamphetamine [MDMA or “ecstasy”]) and nonosmotic stimuli (such as nausea, hypoxia, hypercapnia, and hypoglycemia) can also increase AVP secretion.^1,4–6,23 Previous case reports in dogs have associated SIADH with amebic meningoencephalitis, heartworm disease, hypothalamic neoplasia, and idiopathic etiologies.^7–11

Although the cause of this cat’s SIADH cannot be conclusively determined, we believe it was likely induced by a combination of anesthesia, the surgical procedure, and metoclopramide administration. Anesthesia and surgery have been shown to increase AVP secretion in humans for up to 5 days following surgery, and postoperative hyponatremia has been reported in 4.4% of adults undergoing elective procedures.^26–28 This reaction is seen more often following major surgical procedures; however, SIADH has been reported following an uncomplicated laparoscopic cholecystectomy (a procedure not unlike that performed in the cat of this study).²⁹ Secretion of AVP has been shown to increase in healthy beagles undergoing surgery, but SIADH-induced hyponatremia in an animal following surgery and/or anesthesia has not been reported.³⁰

Many drugs can affect the secretion of AVP, including metoclopramide, which was administered 24 hours prior to clinical decline in the cat reported herein.¹⁶ Plasma AVP concentrations in healthy adult humans have been shown to increase 10 to 20 minutes following IV injection of metoclopramide, and decreased free-water excretion was observed in patients undergoing diuresis.¹⁶ The mechanism for this is not well understood and is thought to be unrelated to the drug’s antidopaminergic properties, as other dopamine antagonists have not been shown to have the same effect.¹⁶ Clinical SIADH has not been reported in association with metoclopramide administration in any species, and one study indicated that metoclopramide alone did not lead to hyponatremia or water retention in humans.³¹ In this case, metoclopramide was possibly a contributing factor. However, concurrent water restriction and discontinuation of metoclopramide make it impossible to establish the role of the drug in this cat’s SIADH.

Vomiting and nausea, as evidenced by ptyalism, were noted in this cat throughout its stay in the hospital and may have also played a role in the development of SIADH. Nausea is a potent stimulus for AVP secretion.^5,23 However, ptyalism and intermittent vomiting continued during the cat’s recovery, making nausea unlikely as the only factor contributing to AVP secretion.

Other possible causes of SIADH include an unidentified malignancy or neurological disease. Without resolution of these conditions, the hyponatremia would have been expected to recur once fluid restrictions were lifted. Thoracic radiographs were not obtained; therefore, occult pulmonary disease cannot be excluded. However, no clinical signs were observed to suggest respiratory pathology.

Clinical signs associated with SIADH are those attributable to hyponatremia and may include weakness, lethargy, anorexia, nausea, vomiting, and seizures as a result of cerebral edema.^3,6,24 Polydipsia has also been reported in dogs with the condition.^8,9 The severity of signs is often related to the rapidity of onset and the amount of time the brain has to compensate for the changes in osmolality; thus, chronic cases may be relatively asymptomatic.^1,6,22 The cat described herein developed severe lethargy and weakness and was unable to walk more than a few steps without assistance. Given the concurrent hypokalemia, the deficits in both sodium and potassium were likely contributing to neuromuscular weakness. Nausea, vomiting, and anorexia were also observed, though these may have been associated with the cat’s liver disease. All of these signs resolved once the cat’s electrolyte disturbances resolved.

The mainstay of treatment for SIADH involves removing any identifiable underlying cause and fluid restriction to allow urine water loss to exceed intake.^6,24 Induction of a negative water balance allows the normal body fluid volume to be restored and urinary sodium excretion to decrease, leading to an increase in serum sodium concentration.²⁴ In cases of chronic hyponatremia, this must be done gradually to avoid brain injury and myelinolysis.^1,4,6 General recommendations suggest that serum sodium concentrations should be increased no more than 10 to 12 mEq/L per day.^1,3 In acute cases, or those with severe clinical signs, treatment with hypertonic (3%) saline with or without the addition of furosemide can be given until the serum sodium concentration reaches 125 mEq/L, at which point fluid restriction should be instituted.^1,4–6,24 While this treatment option is well described in the human literature, its use in animals with SIADH has not been evaluated. The use of demeclocycline, phenytoin, or AVP antagonists such as conivaptan, tolvaptan, and OPC-31260 have also been described in the treatment of human SIADH.^{1,4–6,22,24} Their use in veterinary medicine is limited to one report describing the successful treatment of a dog with SIADH using 3 mg/kg of OPC-31260 q 12 hours.¹⁰

Fluid restriction was successful in reversing this cat’s SIADH. In retrospect, more aggressive fluid restriction early in the course of treatment may have allowed for a more rapid recovery. Though concern existed about the development of myelinolysis, the cat developed hyponatremia acutely, and the fluid intake was above maintenance even after water restrictions were put in place. Following initial therapy, the plasma sodium concentration rose no more than 4 to 5 mEq/L per day, which was well within the recommended range.

Other electrolyte and acid-base disturbances that are not typical of SIADH were also noted in the cat of this report. The cat was diagnosed with a metabolic acidosis, which has not been described in humans with this condition. Although the PCO₂ was normal, this was not considered a mixed acidosis, based on information suggesting that cats show poor, if any, respiratory compensation for metabolic acidosis.³² However, we cannot exclude a respiratory component to the acidosis, which could be caused by hypoventilation associated with the electrolyte disturbances or hepatic encephalopathy. Calculation of the chloride gap and chloride:sodium ratio along with a normal anion gap were consistent with a hyperchloremic acidosis.³³ A potential cause in this cat would be a dilutional acidosis associated with administration of an alkali-free, chloride-containing fluid alone or in combination with the SIADH.³² Distal renal tubular acidosis may also have contributed to the metabolic acidosis as well as the hypokalemia. The neutral urine pH measured during this time period is supportive [Table 2], and distal renal tubular acidosis has been reported in a cat with hepatic lipidosis.³⁴ However, more aggressive alkali therapy is usually necessary to correct such a disorder.³²

Hypokalemia is not a typical finding in SIADH cases, but it has been reported in both humans and a dog with the condition.^7,21,35 In these cases, the hypokalemia has been attributed to concurrent hypomagnesemia, causing intra-cellular loss and decreased potassium reabsorption by the kidneys, or increased aldosterone secretion induced by hypotonicity.^21,35 Calculation of the transtubular potassium gradient suggests aldosterone was contributing to the hypokalemia, lending support to the latter mechanism.¹³ Other potential causes would include distal renal tubular acidosis (although, as discussed earlier, this would appear to have resolved without treatment) and refeeding syndrome, although the elevated fractional excretion of potassium suggests renal loss of the electrolyte. Increased delivery of sodium to the distal nephrons leads to increased activity of sodium-potassium adenosine triphosphatase. This activity, in combination with high tubular flow rates, results in an enhanced concentration gradient for potassium secretion and increased urinary excretion of potassium, which also may have contributed to the hypokalemia.¹³

Conclusion

Syndrome of inappropriate antidiuretic hormone secretion is an uncommon cause of hyponatremia in animals, but it should be considered in hyponatremia cases that are unresponsive to fluid supplementation and that have normal renal, thyroid, and adrenal function. Evaluation of hyponatremia using plasma and urine osmolality and measurement of fractional excretion of sodium can allow for early recognition of the syndrome so appropriate therapy can be instituted. Treatment should be undertaken cautiously and monitored closely to prevent neurological complications of myelinolysis.

Acknowledgments

The authors thank Dr. David Panciera for his help in the preparation of this manuscript.