Electrolyte Disturbances and Cardiac Arrhythmias in a Dog Following Pamidronate, Calcitonin, and Furosemide Administration for Hypercalcemia of Malignancy

Case Report

A 13.5-year-old, 23.6-kg, castrated male Doberman pinscher mixed-breed dog was referred to the Tufts University School of Veterinary Medicine for treatment of a perianal tumor. The owner reported that the dog had episodic urinary incontinence and progressive difficulty in rising and climbing stairs over the previous 4 months. The primary care veterinarian evaluated the dog 2 months prior to hospital admission. Hypercalcemia (13.0 mg/dL; reference range, 8.9 to 12.0 mg/dL) and a slight elevation of creatinine (1.4 mg/dL; reference range, 0.4 to 1.3 mg/dL) were identified at that time. Selegilene^a was prescribed and resulted in some initial clinical improvement. The urinary incontinence and hind-limb weakness subsequently returned, and the dog became anorexic. Four days before referral, repeat laboratory testing documented elevations in total protein (8.8 g/dL; reference range, 5.6 to 8.2 g/dL), total calcium (>15 mg/dL), and creatinine (2.8 mg/dL). The owner sought a second opinion, and the second veterinarian detected a 5-cm mass causing visible deformation to the right side of the anus. The dog was referred for further diagnostic evaluation and treatment.

The owner reported no additional historical problems or prior medical conditions at the time of referral. The dog ate a commercial diet, was fully vaccinated, and was treated monthly with a heartworm preventative medication. On physical examination, the dog was weak, unable to rise, and unable to support weight on the hind limbs. The body condition score was 2/9. Oral mucous membranes were injected and tacky, with a capillary refill time of 2 seconds. Motor and sensory functions were present in all four limbs, and hind-limb reflexes were intact. A metabolic cause for the hind-limb weakness was suspected. The anus was erythematous, and two palpable masses (5 cm and 3 cm) caused visible deformation of the perianal region at the 12 o’ clock and 2 o’ clock positions, respectively. The heart rate was 120 beats per minute (bpm), and no cardiac murmurs or arrhythmias were noted. The respiratory rate and effort, rectal temperature, and abdominal palpation were unremarkable. The patient was admitted to the hospital for further diagnostic testing and treatment.

Pertinent serum biochemical results obtained at the time of admission are reported in the Table. Additional biochemical abnormalities included elevations of total bilirubin (0.62 mg/dL; reference range, 0.10 to 0.50 mg/dL), aspartate aminotransferase (AST, 307; reference range, 15 to 52 U/L), and amylase (2,301 U/L; reference range, 400 to 1,200 U/L). No abnormalities were noted on the complete blood count (CBC); the packed cell volume (PCV) on admission was 42% (reference range, 37% to 55%), and total solids were 7.0 g/dL. The urine specific gravity was 1.013, and there were 0 to 5 red blood cells per high-power field (hpf). Dehydration and sustained hypercalcemia, attributable to malignancy of the perianal tumors, were judged to be contributing factors to the azotemia.

Initial therapy was directed toward the correction of dehydration, hypercalcemia, and azotemia; it consisted of intravenous (IV) 0.9% sodium chloride (NaCl) (6.4 mL/kg body weight per hour), furosemide (2.75 mg/kg body weight, IV q 6 hours), cimetidine (6.3 mg/kg body weight, IV q 8 hours), and calcitonin (7 IU/kg body weight, IV q 12 hours). A urinary catheter and closed collection system were placed, and urine output was recorded every 4 hours. The dog maintained a urinary output >2 to 5 mL/kg body weight per hour throughout hospitalization. Hypercalcemia persisted, and at 21 hours postadmission, pamidronate^b (1.3 mg/kg body weight) was reconstituted, added to 250 mL 0.9% NaCl, and administered IV over 2 hours. Furosemide therapy was discontinued at this time. Intravenous fluid therapy with 0.9% NaCl was continued, and supplemental potassium chloride (KCl) was added to the fluids based on the results of serial serum potassium measurements. Calcitonin and supportive care were continued over the following 48 hours.

On day 2, a fine-needle aspirate (FNA) of the mass was obtained, and thoracic radiographs were taken. The FNA was interpreted as an apocrine gland adenocarcinoma. Thoracic radiographs were unremarkable. The dog improved clinically; serum total calcium concentration declined [see Table]; and signs partially attributable to hypercalcemia (i.e., weakness, lethargy, and anorexia) abated during that time period.

On day 4 of hospitalization (3 days postpamidronate therapy), serum ionized calcium was still mildly elevated (1.49 mmol/L; reference range, 1.29 to 1.33 mmol/L), albumin was 2.5 g/dL, and creatinine decreased to 2.4 mg/dL. The drop in serum albumin was likely a result of prolonged anorexia and resolution of dehydration following fluid therapy. Phosphorous and ionized magnesium concentrations were below reference ranges, but clinical signs referable to hypophosphatemia and hypomagnesemia were not identified, and corrective steps were not initiated. Based on progressive improvement of the dog’s condition and improving renal function, the dog was deemed stable for anesthesia and surgical removal of the perianal mass. The premedication consisted of oxymorphone (0.06 mg/kg body weight, IV) and diazepam (0.1 mg/kg body weight, IV). Anesthesia was induced 10 minutes after premedication with propofol (4.9 mg/kg body weight, IV). The dog was intubated, positioned in sternal recumbency, and anesthesia was maintained with oxygen and isoflurane. Lactated Ringer’s solution was administered (10 mL/kg body weight per minute) during surgery. Monitoring equipment included continuous electrocardiography (ECG), indirect blood pressure measurement, an esophageal temperature probe, esophageal stethoscope, pulse oximetry, and end-tidal carbon dioxide (CO₂). The dog was ventilated at 10 breaths per minute, with a tidal volume adjusted (15 to 20 mL/kg body weight per breath) to obtain an end-tidal CO₂ of <40 mm Hg and an oxygen saturation of >96%. Glycopyrrolate (0.009 mg/kg body weight, intramuscularly [IM]) was administered when the heart rate, during sinus rhythm, was ≤72 bpm. Ten minutes after glycopyrrolate administration, hypotension developed with a mean arterial blood pressure (MAP) of <60 mm Hg. A 200-mL IV bolus of lactated Ringer’s solution was administered, and a constant-rate infusion (CRI) of dopamine (6 μg/kg body weight per minute) was administered. Within 15 minutes, MAP increased to >70 mm Hg. The MAP was maintained between 80 and 125 mm Hg throughout the remainder of anesthesia. Ten minutes after the initiation of the dopamine infusion, a variety of cardiac arrhythmias was recorded sequentially [Figures 1 to 8]. While these arrhythmias were a cause for concern, they did not result in hemodynamic instability. The type of arrhythmia continued to change as various therapies and interventions were considered. Once atrial fibrillation developed and the ventricular rate had risen to 120 bpm [Figure 6], the dopamine CRI was reduced to 2 μg/kg body weight per minute. The dog spontaneously converted to normal sinus rhythm [Figure 8] after the completion of surgery, approximately 5 minutes after discontinuing inhalant anesthesia.

The recovery from anesthesia was smooth and uneventful. Hypomagnesemia was treated postoperatively with the addition of 24 mEq of magnesium sulfate^c to 250 mL of 0.9% NaCl as a CRI at 10 mL per hour (1 mEq/kg body weight q 24 hours). Hydromorphone was administered as an analgesic, and the dog was eating and drinking the day following surgery. No further cardiac arrhythmias were recorded during the subsequent 48 hours of continuous ECG monitoring. Financial constraints precluded repeated measurement of magnesium and other electrolytes in the postoperative period. Mild hypomagnesemia, hypocalcemia, and hypophosphatemia were present on day 7. The dog appeared fully recovered by the fourth postoperative day (day 8) and was discharged from the hospital.

At suture removal 10 days postoperatively, the dog was doing well. Histopathological findings were diagnostic for adenocarcinoma of the apocrine gland of the anal sac. Repeat thoracic radiographs were performed and showed no evidence of metastatic disease. The CBC was within reference ranges, and the only abnormalities on the serum biochemical profile were a mild elevation of creatinine (1.8 mg/dL; reference range, 0.5 to 1.5 mg/dL) and hypomagnesemia (total, 1.2 mEq/L; reference range, 1.4 to 2.7 mEq/L). The owner elected not to pursue adjunctive radiation or chemotherapy. Total magnesium was within reference ranges (1.9 mEq/L) at the 4-month recheck examination.

Discussion

The prevalence of hypercalcemia in dogs with apocrine gland adenocarcinoma is between 25% and 50%.¹² The marked hypercalcemia in the case described above was manifest as weakness of the hind limbs, anorexia, and renal impairment. Hypercalcemia was likely due to humoral hypercalcemia of perianal apocrine adenocarcinoma;³ however, parathyroid hormone-related protein was not measured in this case.

Definitive treatment of hypercalcemia of malignancy requires correction of the underlying cause. Treatment of malignancy-associated hypercalcemia requires volume expansion and diuresis with 0.9% NaCl, and loop diuretics are needed to increase renal calcium excretion. These treatments are generally sufficient in mild cases of hypercalcemia, but more severe cases of hypercalcemia may require additional therapy aimed at inhibiting bone resorption (e.g., calcitonin, bisphosphonates) or decreasing intestinal calcium absorption (e.g., glucocorticoids, calcium-restricted diet). The use of glucocorticoids in cases of hypercalcemia of malignancy is usually not recommended unless they are part of the chemotherapeutic regimen.⁴ Calcitonin has a rapid onset of action but induces only a moderate decrease in serum calcium level,⁵ and patients may develop resistance to the drug after several days.⁴⁶ Calcitonin decreases renal tubular reabsorption of magnesium and other electrolytes, and furosemide increases renal excretion of magnesium together with other electrolytes, so these drugs may have contributed to the electrolyte abnormalities in the dog of this report.⁷ A reduction in serum calcium concentration was necessary in order to correct clinical signs and stabilize the dog for surgical removal of the apocrine gland adenocarcinoma.

Bisphosphonates are inhibitors of normal and pathological bone resorption.⁸⁹ Bisphosphonates reduce serum calcium by inhibiting osteoclastic bone resorption, retarding the deposition of hydroxyapatite in bone collagen, increasing unmineralized osteoid, and inhibiting the formation of calcium phosphate crystals.^8–10 In cases of severe or refractory hypercalcemia of malignancy in human patients, bisphosphonates, such as pamidronate, are effective in normalizing serum calcium levels.⁵¹⁰ Pamidronate and diuresis have been suggested as first-line treatments in childhood and adult hypercalcemia of malignancy.^7–11 Decreases in serum calcium concentration are generally observed 24 to 28 hours postpamidronate administration.⁵

Studies of pamidronate (0.65 to 2.0 mg/kg body weight, IV) in dogs have shown efficacy in reducing hypercalcemia associated with vitamin D₃ and cholecalciferol toxicosis,¹²¹³ including attenuating the adverse renal effects of hypercalcemia (i.e., mineralization and azotemia).¹² In this case, due to the severity of the hypercalcemia and resultant clinical manifestations (i.e., anorexia, weakness, impaired renal function), pamidronate was used in conjunction with saline diuresis, furosemide, and calcitonin to more quickly resolve hypercalcemia.

Electrolyte abnormalities are recognized complications of bisphosphonate treatment in humans.^5101114 Hypocalcemia, hypomagnesemia, hypophosphatemia, and hypokalemia have been documented with pamidronate use in hypercalcemic human patients.^5101415 These electrolyte disturbances occur with greater frequency in patients with preexisting renal disease or volume contraction.¹⁶ A drop in serum phosphorous concentration is a recognized effect of bisphosphonates.⁵⁸¹¹¹⁴ Hypophosphatemia developed following pamidronate treatment in 11 of 30 human patients with normal baseline phosphorus and in 16 of 20 patients with a low baseline phosphorus, although the side effects in this study were judged to be mild.⁵ Three of five children with cancer-related hypercalcemia who were treated with pamidronate developed hypophosphatemia that was severe enough to require treatment with supplemental phosphate.⁸ Hypomagnesemia is reported in 4% to 15% of human patients suffering malignancy-associated hypercalcemia postpamidronate therapy.⁸¹¹¹⁵ Hypomagnesemia was noted in 40% of 50 patients postpamidronate therapy for hypercalcemia of malignancy (25% were hypomagnesemic prior to treatment) in one human study.⁵ One proposed mechanism for hypomagnesemia in humans is renal wasting of magnesium due to hypophosphatemia, which is known to produce a marked increase in renal magnesium excretion.¹⁴ Hypomagnesemia was initially documented prior to hypophosphatemia in the dog of this report and likely was caused by furosemide and diuresis. Other proposed mechanisms for magnesium deficiency following pamidronate treatment include a preexisting total body deficit, diuretic- or calcitonin-induced renal magnesium wasting, or both. Magnesium homeostasis is regulated primarily by intestinal absorption and renal excretion in the dog,¹⁷ and a disruption in renal magnesium handling seems likely in this dog.

The effects and pharmacokinetics of pamidronate in humans with cancer, including those with impaired renal function, have been evaluated.¹⁶ While a transient reduction in renal function was observed in eight of 31 humans with underlying renal insufficiency, hypercalcemia resolved or improved in 91% of patients, and the lack of adverse reactions in this patient population led to recommendations for its use in the treatment of severe hypercalcemia in patients with impaired renal function.¹⁶ Despite the apparent safety of using pamidronate in humans, manufacturer recommendations still include giving the drug in conjunction with vigorous saline diuresis and restoring appropriate urine output before administration of pamidronate.¹⁵

Clinical effects of magnesium deficiency in dogs include refractory hypokalemia or hypocalcemia, muscular weakness, neurological signs, gastrointestinal ileus or anorexia, and cardiac arrhythmias.^17–20 Clinically apparent hypomagnesemia is uncommon in human hypercalcemic patients following pamidronate treatment.⁸⁹ It has been proposed that clinically relevant hypomagnesemia exists when total serum concentration is <1.2 mg/dL in the dog.¹⁷ Serum magnesium comprises approximately 1% of total body magnesium and may be an inaccurate reflection of true tissue magnesium content.¹⁷ The physiologically active ionized form represents 70% of serum magnesium and is proposed to more accurately reflect intracellular ionized magnesium status.¹⁷ Given this patient’s history of anorexia and the relationship between dietary intake of magnesium and total serum magnesium levels, the authors cannot exclude the possibility that there was preexisting total body depletion of magnesium despite the initial total serum magnesium concentration being within the reference range.

Hypomagnesemia may have been a contributing factor in the observed cardiac arrhythmias during anesthesia. Magnesium is an essential cofactor for the membrane-bound sodium/potassium adenosine triphosphate (Na⁺-K⁺ ATPase) pump. Magnesium deficiency inhibits function of the Na⁺-K⁺ ATPase pump and causes a decrease in intracellular potassium concentration and a rise in intracellular sodium concentrations.²¹²² The resultant alterations in the transmembrane potential can lead to cardiac arrhythmia formation.¹⁷²¹²² Hypomagnesemia also leads to blockade of the voltage-dependent K⁺ channels which interferes with cellular repolarization and action-potential propagation.²³ Magnesium deficiency in the dog causes a reduction in the threshold for epinephrine-induced ventricular arrhythmias, and magnesium supplementation can acutely reverse this phenomenon.¹⁸ Arrhythmias implicated by magnesium-deficient states include atrial fibrillation, supraventricular tachycardia, ventricular premature depolarizations, and ventricular tachycardia.^17–1824 While the authors cannot exclude the possibility that arrhythmias arose from interactions between inhalant anesthetics, glycopyrrolate, and dopamine alone, it is likely that the combination of these factors plus hypomagnesemia led to conditions that facilitated arrhythmia formation. Since the arrhythmias resolved prior to correction of the hypomagnesemia, it is unlikely that hypomagnesemia was the sole predisposing factor for arrhythmias. Although glycopyrrolate administration is recommended to counter anesthesia-induced bradycardia, anti-cholinergics by themselves can also induce arrhythmias. Dopamine administration is recommended for hypotension; however, it can predispose to arrhythmias in dose-dependent fashion. It is possible that dopamine may have contributed to the development of atrial fibrillation in this dog. While anesthetized patients commonly require anticholinergetic drugs and inotropic support, most do not experience the marked arrhythmias noted in this patient. Cardiac arrhythmias following pamidronate administration are not reported in the human literature; however, one case of prolonged Q-T interval postpamidronate has been documented.²⁵ Prolongation of the QT interval in this dog may have resulted from the electrolyte abnormalities, or it may have resulted from bradycardia as the QT duration is inversely related to heart rate. The QT interval at slow heart rates was longer than expected using one correction formula,²⁶ and this provides one argument that the slow heart rate was not the sole cause of QT prolongation. However, the QT interval was of expected duration when the heart rate rose to 120 bpm.

In this case, magnesium was supplemented following surgery; however, it would have been advisable to correct magnesium levels prior to anesthesia. Treatment of hypo-magnesemia prior to surgery may have prevented the arrhythmias that occurred during anesthesia. Unfortunately, the degree of hypomagnesemia and the possible relation to the arrhythmias were not recognized until the arrhythmia had resolved. Oral magnesium was not prescribed at the time of hospital discharge, as anorexia had resolved and furosemide and saline diuresis (likely contributing factors to hypomagnesemia) had been discontinued. Oral magnesium supplementation at the time of hospital discharge might have quickened the return of magnesium to the reference range.

Conclusion

In the dog of this report, the combination of pamidronate, saline diuresis, calcitonin, and furosemide was effective in lowering serum calcium and creatinine levels but also led to significant alterations in other serum electrolytes. Electrolyte abnormalities should be anticipated in animals that require aggressive management of hypercalcemia. In particular, these treatments can lead to clinically important hypomagnesemia, which may contribute to arrhythmia formation. Hypomagnesemia should be identified and corrected when pamidronate is administered to treat canine hypercalcemia.