Hypercalcemia Due to Iatrogenic Secondary Hypoadrenocorticism and Diabetes Mellitus in a Cat

Case Report

A 9-year-old, 2.4-kg, spayed female domestic shorthair cat presented with primary complaints of hair loss, polydipsia, polyphagia, and weight loss. The cat had a 2-year history of progressively worsening, generalized pruritus and self-trauma. The cat had received three doses of methylprednisolone acetate at 6.25 mg/kg body weight, intramuscularly (IM) in the preceding 4 months. The most recent dose was administered 30 days prior to presentation.

On presentation, the cat was bright and alert but was underweight with a mildly distended abdomen. The hair coat was generally thin, with alopecia on the ventral abdomen. The skin was inelastic and extremely thin, with small blood vessels easily visualized. Erythema and scaling were noted in some locations. Deep and superficial skin scrapings were negative for parasites, and cytopathology from alopecic areas revealed no inflammatory cells or bacteria. Blood and urine samples were acquired for analysis with extreme difficulty, resulting in a highly stressful event for the patient. The patient was discharged with the laboratory results pending. Abnormalities on the complete blood count were limited to lymphopenia of 200 cells/μL (reference range, 2.0 to 7.0 × 10³/μL). Abnormalities on the serum biochemical profile^a were limited to mild hypochloridemia of 112 mmol/L (reference range, 116 to 130 mmol/L), hyperglycemia of 526 mg/dL (reference range, 63 to 140 mg/dL), and mildly elevated alanine transferase (ALT) of 145 U/L (reference range, 27 to 127 U/L). Serum total calcium was within reference range at 11.1 mg/dL (reference range, 8.2 to 11.5 mg/dL), and phosphorus was 5.0 mg/dL (reference range, 2.7 to 6.5 mg/dL). Serum total thyroxine (T₄) was within reference range at 17.9 nmol/L (reference range, 10.0 to 45.5 nmol/L). Urinalysis indicated a moderately concentrated (specific gravity, 1.033), acidic (pH, 6.5) urine with glycosuria (2,000 mg/dL), proteinuria (30 mg/dL), and an unremarkable sediment. Urine culture was negative. Serum fructosamine was elevated at 460 μmol/L (reference range, 175 to 400 μmol/L). Based on these results, diabetes mellitus, probably associated with corticosteroid administration, was diagnosed, and plans were made to initiate insulin therapy on the next day.

The following day, the owner noted that the cat was depressed, had decreased appetite, and had vomited several times. She was presented for reevaluation approximately 28 hours after the previous visit. On examination, the cat was extremely depressed and weak, had lost 0.4-kg body weight as compared to the previous day, and was approximately 15% dehydrated. Available laboratory analyses were limited because of the late hour at the time of presentation. Packed cell volume (PCV, 36%; reference range, 24% to 45%) and serum total solids (TS, 10.2 g/dL; reference range, 6.0 to 7.5 g/dL) had both increased approximately 10% in the proceeding 28 hours, supporting marked dehydration. Serum glucose was greater than 500 mg/dL.^b Urine dipstick evaluation of voided urine indicated continued glycosuria (2,000 mg/dL) but no ketonuria. Serum urea nitrogen was within reference range at 26.6 mg/dL (reference range, 16.0 to 36.0 mg/dL), but albumin was elevated at 4.40 g/dL (reference range, 2.60 to 3.90 g/dL).^c Whole-blood electrolyte analysis^d revealed hypernatremia of 166 nmol/L (reference range, 148 to 160 nmol/L), normal serum potassium of 4.3 nmol/L (reference range, 3.66 to 5.70 nmol/L), normal chloride of 121 mmol/L (reference range, 116 to 127 nmol/L), and marked hypercalcemia with an ionized calcium of 7.75 mmol/L (reference range, 4.28 to 5.76 mmol/L). Serum total calcium^c was elevated at greater than 16.0 mg/dL (reference range, 7.8 to 11.3 mg/dL), but the sample quantity was insufficient for dilution and reevaluation.

The cat was managed with intravenous fluid therapy (0.9% sodium chloride) administered at 16 mL/kg body weight per hour for 6 hours, then 5 mL/kg body weight per hour for 6 hours. Regular insulin (0.5 U/kg body weight) was administered IM every 6 hours. By the following morning, the cat was clinically well hydrated and had gained 0.3 kg in body weight. She was bright and alert and willing to eat and drink. Abnormalities on a complete serum biochemistry profile^a were hyperglycemia (298 mg/dL), hypercalcemia (13.4 mg/dL), and slightly elevated ALT (157 U/L). Ionized calcium^d was also elevated (6.92 mg/dL), but serum phosphorus was within reference range (4.3 mg/dL).

Cytopathological examination of fine-needle aspirates of two peripheral lymph nodes yielded normal lymphoid tissue. Thoracic radiographs were unremarkable. A bone-marrow aspirate indicated normal erythro- and myelopoiesis and failed to identify any evidence for malignancy. Ultra-sonographic examination of the abdomen failed to identify any abnormalities consistent with occult neoplasia. A 3-mm diameter cyst was identified in the left kidney. The cyst had been visualized on a previous ultrasound examination performed 2 years earlier. Ultrasonographic examination of the ventral neck failed to identify any parathyroid masses or enlargement. Hypercalcemia secondary to toxicosis was ruled out because of the lack of potential exposure, as she was a strictly indoor cat. An adrenocorticotropic hormone (ACTH) stimulation test^e indicated diminished adrenal response, with a preACTH serum cortisol level of 17.5 nmol/L (reference range, 28 to 110 nmol/L) and a 1-hour postACTH serum cortisol level of 63.3 nmol/L (reference range, 172 to 464 nmol/L).

The cat was discharged with Humulin Ultralente insulin at 0.5 U/kg body weight, subcutaneously (SC) once daily, and stress-dose prednisone at 0.5 mg/kg body weight orally once daily. The following day the cat was approximately 5% dehydrated on physical examination. Ionized calcium^d was 6.21 mg/dL. Physiological saline was dispensed for the owner to administer at a rate of 20 mL/kg body weight SC daily. On day 6, the cat was again approximately 5% dehydrated, and the ionized calcium^d was within reference range at 5.41 mg/dL. The owner increased the quantity of SC fluids administered to 40 mL/kg body weight per day.

Over the next week, the polydipsia, polyuria, and polyphagia persisted, indicating poor control of the diabetes mellitus. Periodic evaluation of whole-blood glucose^b at home by the owner indicated constant hyperglycemia. A recheck ionized calcium^d on day 13 was 5.21 mg/dL. Over the next several weeks, the Ultralente insulin dose was increased to 1 U/kg body weight and then to 1.5 U/kg body weight without clinical improvement. At the same time, the oral prednisone supplementation was gradually decreased to 0.2 mg/kg body weight daily. Due to lack of adequate control with Ultralente insulin, the insulin protocol was changed on day 21 to beef/pork isophane (NPH) insulin at 1.5 U/kg body weight SC twice daily. Within 3 days, the polydipsia, polyuria, and polyphagia resolved. The cat was clinically normal other than the return of the pruritus for the next several weeks. Periodic at-home glucose evaluations^b by the owner indicated some late-afternoon hypoglycemia. The insulin dose was halved and then discontinued altogether 36 days after initial presentation (66 days since administration of the most recent dose of methylprednisolone acetate).

Day 40 follow-up evaluation indicated a clinically normal cat, other than the pruritus and alopecia. Hair regrowth was evident on the ventral abdomen. The cat’s current body weight had increased to 2.9 kg. Resolution of the previous diabetes mellitus was documented with a serum glucose of 70 mg/dL (despite no recent insulin administration) and a serum fructosamine of 268 μmol/L. The cat was also normocalcemic with an ionized calcium^d of 4.31 mg/dL and a serum total calcium^a of 9.7 mg/dL. An ACTH stimulation test indicated a normal adrenal response with a preACTH serum cortisol level of 69.1 nmol/L and a 1-hour postACTH serum cortisol level of 188.0 nmol/L. Oral prednisone supplementation was then discontinued. The cat remained clinically normal other than the pruritus for the next 28 months of follow-up.

Discussion

The development of transient diabetes mellitus in cats associated with the use of methylprednisolone acetate and other diabetogenic hormones is well documented.¹ At the time of presentation for polyphagia, polydipsia, and polyuria, this cat appeared to be a stable, uncomplicated diabetic with clinical indications of iatrogenic hyperadrenocorticism. The diabetes mellitus responded poorly to Humulin-recombinant Ultralente insulin, as has been noted in other feline diabetic patients.²³

Cats are generally considered to be resistant to the development of iatrogenic hyperadrenocorticism.⁴ Reports of feline iatrogenic hyperadrenocorticism are rare in the veterinary literature, despite extensive use of repositol corticosteroids for inflammatory diseases. Methylprednisolone acetate given once weekly to normal cats for 4 weeks failed to induce clinical evidence of hyperadrenocorticism, but produced a blunted adrenal response after only one dose of 20 mg per cat.⁵ The same dose given weekly to normal cats produced clinical Cushing’s syndrome after weekly injections of 20 mg per cat for 9 to 11 successive weeks.⁶ The degree of clinical evidence of Cushing’s syndrome (e.g., alopecia, thin skin, abdominal distention) seen in this cat was unusual, given the minimal use of the drug (a total of 45 mg administered over a 4-month period). The blunted adrenal response to administration of ACTH confirmed the suppression of the hypothalamic-pituitary-adrenocortical axis. Apparently adrenal function was adequate while the patient was at home, but the stress of veterinary evaluation and venipuncture precipitated a glucocorticoid-deficient Addisonian crisis, resulting in weakness, depression, and vomiting. The glycosuria and acute-onset hypercalcemia caused excessive renal water loss and secondary profound dehydration. Because of the degree of clinical compromise seen with the adrenal insufficiency crisis and the presence of concurrent illness (i.e., diabetes mellitus), a stress level of glucocorticoid supplementation was used initially. As the patient clinically improved, the dose was reduced to meet physiological glucocorticoid requirements and then was discontinued when the ACTH stimulation test suggested adequate adrenal function.

Mineralocorticoid production is unaffected by the exogenous administration of glucocorticoids. This patient’s adrenal insufficiency was consequently not associated with hyperkalemia, hyponatremia, or the added volume depletion that leads to prerenal azotemia when mineralocorticoid deficiency is present.

The occurrence of hypercalcemia in conjunction with naturally occurring hypoadrenocorticism has been reported in dogs,⁷ humans,^8–10 and one cat.¹¹ Hypoadrenocorticism, whether primary or secondary, is rare in the cat.^611–13 In the previous reports of hypercalcemia with acute adrenal insufficiency in dogs⁷ and in the cat,¹¹ the patients were deficient in both glucocorticoid and mineralocorticoid production (as indicated by the occurrence of hyperkalemia and hyponatremia). The presence of hypercalcemia in dogs was correlated with the severity of the adrenocortical insufficiency, with all dogs presenting with moderate to severe azotemia and evidence of volume depletion.⁷ The finding of dehydration in these dogs may have elevated the measured total calcium levels due to an increase in protein-bound calcium. This phenomenon was noted in experimentally adrenalectomized dogs.¹⁴ Ionized calcium has not been measured in the reported cases of hypoadrenocorticism with hypercalcemia, so it is difficult to assess the true severity of the hypercalcemia previously reported. The degree of hypercalcemia as measured by total calcium has been generally mild to moderate. The total serum calcium ranged from 12.0 to 14.9 mg/dL in dogs with hypoadrenocorticism,⁷ and the single reported cat had a calcium level of 14.0 mg/dL.¹¹ The feline patient in this report had marked hypercalcemia confirmed with both an elevated total calcium of >16.0 mg/dL and an elevated ionized calcium of 7.75 mg/dL. The hypercalcemia could not be explained with dehydration alone, since the ionized calcium was elevated, and the abnormality persisted for several days despite correction of the hydration deficit.

The reduction in renal function secondary to hypovolemia may also have played a role in the development of hypercalcemia, as decreased glomerular filtration rate may be associated with increased tubular reabsorption of calcium. The partial correction of the hypercalcemia noted with volume replacement in this cat was similar to that reported in human patients,⁸¹⁰ and it supports this interpretation. However, in this cat and in reported human cases,⁸ calcium levels did not normalize until glucocorticoid deficiency was also corrected.

The exact mechanism of development of hypercalcemia with hypoadrenocorticism is unclear, but it is likely multi-factorial. Because glucocorticoids antagonize the actions of vitamin D, they decrease intestinal absorption of calcium. Glucocorticoid deficiency may therefore result in excessive intestinal calcium absorption. However, experimental studies of calcium absorption in adrenalectomized dogs failed to support this hypothesis.¹⁵ The peracute onset of the hypercalcemia in this patient, despite lack of food intake, would also suggest that the effect is not mediated by increased activity of vitamin D. Human cases have demonstrated a suppression of the parathyroid hormone (PTH)-vitamin D axis.¹⁰ The authors were not able to measure intact PTH concentration in this cat because of a lack of a validated feline PTH assay.

Decreased renal excretion of calcium may also contribute to the hypercalcemia of glucocorticoid deficiency. It has been suggested that the rising serum calcium level interferes with distal renal tubular sodium and water retention. The increased sodium reabsorption that then occurs in the proximal tubule is associated with increased calcium reabsorption.⁸

Clinical and experimental evidence is most suggestive that the hypercalcemia of glucocorticoid deficiency is in large part dependent on increased resorption of calcium from bone. Hypercalcemic adrenal-deficient human patients have developed increased urinary markers of bone resorption,⁹ but bone biopsies have failed to indicate increased osteoclastic activity.¹⁰ Glucocorticoids may have an impact on the differentiation of osteoblasts through as-of-yet unidentified mechanisms.¹⁰ Thyroid hormone is thought to play a role in increased bone calcium mobilization in glucocorticoid deficiency, as thyroid hormone has been shown to stimulate bone resorption in tissue culture¹⁶ and in mice.¹⁷ Studies of adrenalectomized dogs indicate that hypercalcemia occurs only if the thyroid gland is present and functioning,¹⁸ and glucocorticoid-deficient hypercalcemic crises may be more likely in humans with thyroid diseases.⁸ This cat’s thyroid status was normal, indicating that hyperthyroidism is not a prerequisite for development of hypercalcemia with hypoadrenocorticism in cats.

Because this cat was normocalcemic immediately prior to onset of the glucocorticoid deficiency crisis, clearly the mechanism of development of the hypercalcemia involves a rapid mobilization of calcium from body stores, an acute change in the level of renal calcium excretion, or both. Glucocorticoid deficiency should be considered a differential diagnosis for hypercalcemia in cats.