A case of hyperglycemic hyperosmolar syndrome in a pug dog

Abstract

An 8-year-old spayed female pug dog that had recently been diagnosed with pancreatitis was brought to a veterinary clinic with polyuria, polydipsia, and seizures. The dog was diagnosed with hyperglycemic hyperosmolar syndrome, a rare complication of diabetes mellitus. The dog was hospitalized and was administered intravenous fluids and insulin therapy, and clinical signs improved.

On May 10, 2022, a spayed 8-year-old female pug dog was brought to the Dawson Creek Veterinary Clinic in Dawson Creek, British Columbia. The dog had exhibited clinical signs of vomiting, diarrhea, and anorexia for approximately 1 wk. The pug was up-to-date on rabies, canine distemper virus, canine adenovirus type-2, and canine parvovirus vaccinations, and had no travel history or known history of dietary indiscretion. On presentation, the dog was clinically stable and there were no abnormal findings on physical examination. A complete blood (cell) count (CBC) revealed a moderate leukocytosis [21.19 × 10⁹/L; reference interval (RI): 6.00 to 17.00 × 10⁹/L] characterized by a moderate neutrophilia (19.71 × 10⁹/L; RI: 3.62 to 12.30 × 10⁹/L) and a mild lymphopenia (0.79 × 10⁹; RI: 0.83 to 4.91 × 10⁹/L). Biochemistry revealed a mild low blood urea nitrogen (BUN) (3.15 mmol/L; RI: 3.21 to 10.35 mmol/L), mild hyperglobulinemia (43 g/L; RI: 20 to 36 g/L), mild hyperglycemia (8.5 mmol/L; RI: 4.2 to 6.9 mmol/L), and a very high alkaline phosphatase (ALP) value (609 U/L; RI: 0 to 140 U/L). The total T4 was low (8.44 nmol/L; RI: 15.40 to 55.30 nmol/L), symmetric dimethylarginine (SDMA) was within normal range (8.0 μg/dL; RI: 0.0 to 15.0 μg/dL), and 1,2-o-dilauryl-rac-glycero-3-glutaric acid-(6′-methylresorufin) ester (DGGR) lipase was very high (300.0 U/L; RI: 0.0 to 125.0 U/L). The dog was diagnosed with pancreatitis and the owners opted to take her home for continued monitoring with supportive therapy [tramadol HCl compounded; Summit Veterinary Pharmacy, Aurora, Ontario), 50 mg, PO, q12h for 4 d; maropitant citrate (Cerenia; Zoetis, Kirkland, Quebec), 16 mg, PO, q24h for 4 d; marbofloxacin (Zeniquin; Zoetis), 50 mg, PO, q24h for 7 d]. The owners were advised to return in 1 mo to repeat the total T4 test to rule out hypothyroidism (as opposed to euthyroid sick syndrome).

The dog was returned the following day at which time she was 7% dehydrated and slight pain was detected on cranial abdominal palpation. She was treated symptomatically for 24 h with fluid rehydration and other medications [maropitant (Cerenia; Zoetis), 1 mg/kg, IV, once, enrofloxacin (Baytril; Elanco, Guelph, Ontario), 5 mg/kg, IV, once, butorphanol (Torbugesic; Zoetis), 0.2 mg/kg, IV, once, Famotidine (Omega, Montreal, Quebec), 1 mg/kg, IV, once]. She was discharged the following day doing well.

On June 9, the dog was returned to the Dawson Creek Veterinary Clinic with clinical signs of polyuria, polydipsia, and polyphagia of 2 wk duration. There had been no lifestyle changes or known dietary indiscretions since her last visit. The owner reported that the dog had a seizure in the car on the way into the clinic. Upon arrival at the clinic, the dog was in the middle of a generalized tonic-clonic seizure. An IV catheter was immediately placed and 2 mg/kg of diazepam (Sandoz, Boucherville, Quebec) was titrated over the course of 5 min to stop the seizure. The dog did not have a history of seizures or other neurological disorders. The dog was quite agitated and ataxic after the seizure, but it was difficult to differentiate if it was a post-ictal change or due to clinic-related anxiety, of which there was historical evidence. The pug also had a cough and stertorous breathing, which was attributed to her brachycephalic nature.

Urinalysis was run on a free-catch urine sample which revealed glucosuria (1000 mg/dL) and apparently adequately concentrated urine (USG 1.031). Urine dipsticks performed with and without hydrogen peroxide added to the urine were both negative for ketonuria. A CBC was unremarkable, and a biochemistry panel revealed a severe hyperglycemia (> 33.3 mmol/L; RI: 4.2 to 6.9 mmol/L), a severe increase in ALP (596 U/L; RI: 0 to 140 U/L), and a mild hypochloremia (98 mmol/L; RI: 102 to 120 mmol/L). The SDMA was within the normal range (8.9 μg/dL; RI: 0.0 to 15.0 μg/dL). The serum osmolality was calculated as 346.63 mmol/L using the method as described by Randels-Thorp (1). The elevated blood glucose and glucosuria were diagnostic for diabetes mellitus. The above findings also met many of the criteria for hyperglycemic hyperosmolar syndrome (HSS), as described by Randels-Thorp. A notable exception was that profound dehydration was not evident from the clinical findings or laboratory analyses. The possibility of other concurrent illness, such as hyperadrenocorticism or hypothyroidism, had also not been ruled out.

A second IV catheter was placed, and the dog was hospitalized for 25 h. Intravenous 0.9% saline was given at a rate of 2× maintenance initially, and the dog was started on a regular insulin (Humalog; Lilly, Toronto, Ontario) constant rate infusion (CRI) at a rate of 0.04 U/kg per hour. The insulin CRI was created by adding 1 U/kg of regular insulin to a 250-mL bag of 0.9% saline and then running 50 mL through the line before starting the infusion. The dog’s blood glucose was monitored throughout hospitalization and the insulin CRI rate, as well as the regular saline fluid rate, were adjusted accordingly (Table 1). Dextrose 50% (Vétoquinol, Lavaltrie, Quebec) was added to the intravenous fluids when blood glucose levels fell below 14 mmol/L. After 4 h of stabilization, the dog was no longer ataxic and had shown great clinical improvement. Her electrolytes, BUN, and serum osmolality were monitored and calculated throughout the night. After 18 h of hospitalization, the pug was doing well clinically, and the insulin CRI was discontinued and a dose of 0.25 U/kg porcine insulin (Caninsulin; Merck, Kirkland, Quebec) was given subcutaneously. After 25 h of hospitalization, the dog remained stable and was discharged from the hospital with Caninsulin to administer at home (0.25 U/kg caninsulin, q12h, SC). At the time of writing, a follow-up had not yet been performed on the pug.

Table 1.

The dog’s blood glucose curve, insulin constant rate infusion (CRI) rate, hydration fluids, and serum osmolality throughout her stay in hospital from June 9 to 10, 2022. Please note her maintenance fluid rate was 20 mL/h. Porcine insulin (Caninsulin; Merck, Kirkland, Quebec) and Dextrose 50% (Vétoquinol, Lavaltrie, Quebec) were used. The dog was fed Hills Z/D canned food because this is what her owners had been feeding her at home.

Date	Time	Blood glucose concentration (mmol/L)	Insulin CRI (U/kg per hour)	IV hydration fluids	Serum osmolality (mmol/L)	Notes
June 9	14:00	> 33.3	Not yet started	Not yet started	346.63
	16:00	20.3	0.04	0.9% NaCl; 40 mL/h
	17:00	24.5	0.04	0.9% NaCl; 40 mL/h
	18:00	24.5	0.04	0.9% NaCl; 40 mL/h	341.52
	19:00	16.9	0.02	0.9% NaCl; 20 mL/h
	20:00	12.8	0.02	0.9% NaCl + 2.5% dextrose; 20 mL/h		Fed 2 tbsp of Hills Z/D canned food
	21:00	23.2	0.02	0.9% NaCl + 2.5% dextrose; 15 mL/h	334.2
	22:00	16	0.02	0.9% NaCl + 2.5% dextrose; 15 mL/h
	23:00	16.6	0.02	0.9% NaCl + 2.5% dextrose; 15 mL/h
June 10	0:00	15.3	0.02	0.9% NaCl + 2.5% dextrose; 15 mL/h	321.39
	2:00	12.9	0.02	0.9% NaCl + 2.5% dextrose; 15 mL/h
	4:00	14.8	0.02	0.9% NaCl + 2.5% dextrose; 15 mL/h
	6:00	11.3	0.02	0.9% NaCl + 2.5% dextrose; 15 mL/h
	7:00	8.9	0.02	0.9% NaCl + 2.5% dextrose; 15 mL/h		Fed 1/4 can of Hills Z/D canned food
	8:00	10	Stopped CRI; administered 0.25 U/kg caninsulin, SC	0.9% NaCl + 2.5% dextrose; 15 mL/h
	10:00	16.2		0.9% NaCl + 2.5% dextrose; 15 mL/h
	12:00	5.1		0.9% NaCl + 2.5% Dextrose; 15 mL/h
	13:00	6.3		0.9% NaCl + 2.5% Dextrose; 15 mL/h
	14:00	4.5		0.9% NaCl + 2.5% Dextrose; 15 mL/h		Fed 1/4 can of Hills Z/D canned food
	15:00	13.7		Stopped

Discussion

Hyperglycemic hyperosmolar syndrome (HHS) is a rare complication of diabetes mellitus that has been described in humans and domestic animals. In one retrospective study, HHS was diagnosed in 5% of dogs with diabetes mellitus (2). Diabetic ketoacidosis (DKA) and HHS are both extreme complications of diabetes mellitus and are characterized by an absolute or relative deficiency of insulin, and an abundance of counter-regulatory hormones. However, the lack of pronounced ketonemia and/or ketonuria is primarily what distinguishes HHS from DKA (3). The pathophysiology of HHS is still somewhat poorly understood. It is believed that HHS patients have some beta cells in the pancreas that are still able to produce insulin. This small amount of insulin prevents the formation of ketones by preventing lipolysis. Elevated counter-regulatory hormones glucagon, catecholamines, and cortisol) stimulate glycogenolysis, gluconeogenesis, and proteolysis. The combination of increased glucose production and inadequate cellular glucose uptake (due to a deficiency of insulin) leads to hyperglycemia, which causes osmotic diuresis of the urine and leads to dehydration (1). Dehydration results in a decreased glomerular filtration rate due to decreased kidney perfusion. This causes decreased glucose excretion which further exacerbates the hyperglycemia. In cases of HHS, the hyperglycemia typically persists unnoticed for a much longer period than in DKA cases because there is no ketosis and therefore no clinical signs of metabolic acidosis. Prolonged hyperglycemia leads to increased plasma osmolality, which causes water to be drawn out of cerebral neurons. This further decreases water intake and, if severe enough, can alter mental status (3).

The presentation of patients with HHS is often triggered by gastrointestinal signs (such as vomiting and anorexia) and neurologic abnormalities (such as weakness and lethargy). In addition, drugs that alter carbohydrate metabolism (such as corticosteroids, thiazides, sympathomimetic agents, and progesterone) have been known to precipitate the onset of an HHS crisis (4). Dogs with HHS may have been recently diagnosed with diabetes mellitus, and they may have other concurrent diseases such as renal insufficiency, cardiac disease, hyperadrenocorticism, and pancreatitis (1). These animals may show clinical signs of severe dehydration, hypothermia, lethargy, depression, and altered mental status. If hyperosmolality is severe enough, there may be additional neurological signs such as altered pupillary light reflexes (PLRs), circling, pacing, and seizures (1).

The established criteria for diagnosing HHS are: blood glucose > 33.3 mmol/L, serum osmolality ≥ 320 mmol/L, profound dehydration, pH > 7.3, bicarbonate > 15 mEq/L, small ketonuria, absent to low ketonemia, and altered level of consciousness (1). There are often abnormal neurological signs if the animal has a serum osmolality > 340 mmol/L. The dog reported here had a severe hyperglycemia and a calculated osmolality of 346.63 mmol/L, which could have been the cause of her seizures and ataxia. Due to the rural location of the clinic and with the concurrent limited resources, the dog’s blood gases could not be measured, so the blood pH and bicarbonate levels are unknown. However, the dog was not displaying clinical signs of metabolic acidosis, so it was not suspected at the time. The urine dipsticks performed on the pug were both negative for ketones, which is supportive but not conclusive of the absence of ketonuria. The nitroprusside reagent used in urine dipsticks detects only acetoacetate and acetone and is not sensitive for beta-hydroxybutyrate (BHB), which is a precursor and the most prevalent ketone in diabetic animals (5). The addition of hydrogen peroxide to urine in a 1:10 ratio before testing with a dipstick is known to convert some urine BHB into acetoacetate and acetone, which was thought to improve the usefulness of the test. However, more recent studies have reported that the addition of hydrogen peroxide does not improve the reading to a clinically relevant level (5–7). Therefore, although ketonuria was not suspected in the reported dog, it cannot be completely ruled out.

The one finding in the dog reported here that was strongly inconsistent with HHS, was the lack of profound dehydration. Both clinical findings and lab work indicated that the dog was only slightly dehydrated. The author was unable to find any reports of previous HHS cases in human or veterinary medicine in which the patient was not severely dehydrated. It is not currently understood why this dog was not profoundly dehydrated but suggests that further research may be needed in this area.

Treatment of HHS is generally similar to that of DKA. The mainstays of therapy are rehydration, insulin administration, careful monitoring, and electrolyte supplementation as needed. Generally, 0.9% saline is recommended for rehydration and to provide sodium to replace glucose in the extracellular spaces. Insulin therapy is usually performed in a similar manner to that of DKA management. A low-dose, short-acting insulin CRI is generally preferred, with constant monitoring of blood glucose (q1 to 2h) and supplementation with dextrose as needed (Table 2) for gradual blood glucose reduction. Judicious monitoring is required in HHS patients to ensure that plasma osmolality does not drop too quickly, to avoid causing cerebral edema. For this reason, it is recommended to decrease the animal’s plasma osmolality by no more than 10 to 15 mOsm/day (1). This requires close monitoring of blood glucose and electrolytes, particularly sodium. In addition, urine output and renal values should be carefully monitored throughout treatment. Although there can be a hypokalemia due to osmotic diuresis, anorexia, and vomiting, this is usually less severe in HHS than DKA patients. Potassium supplementation should be considered on an as-needed basis (3).

Table 2.

Recommended intravenous fluids, and rate of insulin solution according to patient blood glucose concentrations. This is based on an insulin CRI made using a 250 mL bag of 0.9% saline with 2.2 U/kg of regular insulin added to it. It is suggested to run 50 mL of insulin solution through the administration set before hooking up to the patient because insulin will bind to the IV tubing plastic. This table was adapted from Huang and Scott-Moncrieff (8), but the blood glucose concentration has been converted from mg/dL to mmol/L, to better suit the author’s use.

Blood glucose concentration (mmol/L)	IV Hydration fluids	Rate of administration of insulin solution (mL/h)
> 13.9	0.9% saline	10
11.1 to 13.9	0.9% saline + 2.5% dextrose	7
8.3 to 11.1	0.9% saline + 2.5% dextrose	5
5.6 to 8.3	0.9% saline + 5% dextrose	5
< 5.6	0.9% saline + 5% dextrose	Discontinued

It has been estimated that in 69 to 92% of cases with DKA and HHS there are comorbidities such as infection, pyometra, gastroenteritis, renal or hepatic disease, pancreatitis, hypderadrenocorticism, neoplasia, and/or cardiac disease (4). Hyperadrenocorticism is among the most common of these comorbidities, but it is advised to wait to test adrenal function until the patient has been systemically healthy for at least 2 wk, otherwise a false positive result would be expected (8). At the time of writing, further diagnostic testing on this pug had not been performed. Chronic or acute kidney disease is also often seen in animals with HHS, but blood urea and creatinine can also be elevated due to dehydration, so kidney disease should likewise not be diagnosed until any diabetic crisis has been adequately managed (3).

Animals with HHS often have liver damage as a result of hypovolemia and poor hepatic blood flow. Increased alanine aminotransferase (ALT) and aspartate aminotransferase (AST) are commonly seen, but there can also be an elevation in alkaline phosphatase (ALP) due to concurrent pancreatitis and secondary cholestasis (3). This dog had increased ALP, but ALT was within normal limits, which is supportive of pancreatitis and cholestasis without liver damage. The AST was not measured in this case.

There is a very guarded prognosis for diabetic dogs with HHS. In humans, the mortality rates have varied greatly from approximately 10 to 50% (9,10). Due to common comorbidities, long-term survival in dogs has been reportedly very low. It has been demonstrated that a poorer outcome is associated with dogs which have altered mental status and low venous pH when first seen (2). It is estimated that approximately 12% of cats survive long-term post-treatment. There are not yet any clear data on survival rates in dogs (1). Further research would be beneficial to better understand this syndrome in animals so that practitioners can be well-prepared to encounter it in practice.