Beta Cell Neoplasia and the Hypoglycaemic Crisis
British Small Animal Veterinary Congress 2008
Jane Eastwood, MA, VetMB, DVR, DSAM, MRVCS
The Royal Veterinary College
North Mymms, Hatfield, Hertfordshire

Introduction

Pancreatic beta cell tumours or insulinomas are considered to be uncommon in the dog and rare in the cat, with only five reported feline cases in the literature including a recent case report from 2007. In people the incidence of insulinoma is thought to be one to four new cases per million population per year and the clinical scenario differs as they tend to be benign solitary tumours in 90% of cases. In dogs 95% of insulinomas are considered to be aggressive malignant tumours. While they may appear relatively benign on histopathology with ordered, well differentiated cells and few mitotic figures, this does not correlate well with their frequent metastatic rate. Surgery is still considered to be the single most appropriate treatment but unfortunately at the time of surgical intervention most dogs will already have microscopic or grossly visible evidence of metastatic disease. Insulinoma is therefore a clinical syndrome that we are left to manage rather than cure in the majority of cases.

Pathophysiology

Insulin, produced by the pancreatic beta cells, plays an essential role in the complex, carefully regulated interplay of endocrine agents responsible for glucose homeostasis whereby postprandial hyperglycaemia is blunted and the nervous system, which is highly sensitive to the effects of hypoglycaemia, is protected. Insulin lowers blood glucose by facilitating glucose uptake and suppressing endogenous glucose production through gluconeogenesis and glycogenolysis. A hierarchy of counter-regulatory hormones includes glucagon and catecholamines (crucial in the acute phase of any hypoglycaemic response) and growth hormone and cortisol, which are involved in the response to prolonged insulin-induced hypoglycaemia. Glucose homeostasis also relies on normal hepatic blood supply and adequate numbers of functioning hepatocytes, with the renal cortex also contributing in a minor way to gluconeogenesis. While clinical signs might initially be transient, uncontrolled, autonomous production and episodic release of insulin from neoplastic beta cells ultimately overwhelms the counter-regulatory response resulting in significant, prolonged hypoglycaemia.

Clinical Presentation

Although dogs with insulin-secreting tumours tend to present relatively acutely, careful questioning often reveals that subtle waxing and waning clinical signs (Figure 1) might have preceded that event by some time. The late presentation partly explains why the metastatic rate is so high at the time of diagnosis. Subtle signs due to stimulation of the sympathoadrenal system are more often missed in our patients than in people. The onset of neuroglycopenic signs can be delayed by up-regulation of glucose transporters at the blood- brain barrier as an adaptive response to chronic severe hypoglycaemia. Once these signs develop, however, urgent and effective treatment is mandatory as irreversible brain lesions are known to occur in people and have been reported in a dog and a cat with persistent hypoglycaemia.

Figure 1. Clinical signs associated with insulin-secreting tumours.

Autonomic nervous system signs:
The catecholamine surge

Muscle fasciculation
Nervousness
Restlessness
Hunger

Neuroglycopenic signs

Lethargy
Weakness and episodic collapse
Ataxia
Bizarre behaviour including aggression
Seizures
Coma
Peripheral neuropathy

Differential Diagnoses for Hypoglycaemia

The clinical presentation means any symptomatic dog or cat should immediately have a blood glucose checked as part of patient evaluation. The main differentials for profound, symptomatic hypoglycaemia in an adult patient that is otherwise well include insulinoma, iatrogenic inappropriate administration of insulin, xylitol toxicity and non-pancreatic neoplasia. Paraneoplastic hypoglycaemia occurs in dogs with mesenchymal tumours, such as leiomyosarcoma and leiomyoma of the gastrointestinal (GI) tract and liver or epithelial tumours, including liver carcinoma, as well as other tumours including oral melanoma, haemangiosarcoma and salivary gland adenocarcinoma. While the exact mechanism is not always known and might be multifactorial, in one report of hepatocellular carcinoma in a dog the hypoglycaemia was linked to excessive production of insulin-like growth factor type II (IGF II). In juvenile toy breed dogs, quite profound hypoglycaemia can be an acute clinical problem often associated with some additional stress factor. While sepsis (pyometra, pyothorax, prostatic abscess) and systemic inflammatory response syndrome (SIRS) can be associated with quite severe hypoglycaemia there are usually other clear findings on physical examination to support this differential. Spurious blood glucose results occur as a laboratory artefact due to slow processing of samples if immediate in-house testing is not performed. Although not always accurate at low blood glucose levels, an initial glucometer reading can be more helpful than waiting for external laboratory results.

Other differentials typically associated with less severe hypoglycaemia include liver dysfunction due to portosystemic shunt or severe hepatopathy and hypoadrenocorticism. These patients also tend to have other additional unrelated presenting signs.

Diagnosis

Documentation of Hypoglycaemia

When the hypoglycaemic period has been missed then documentation of hypoglycaemia might require a prolonged fast, exercise or sometimes feeding to precipitate excessive insulin release. Low fructosamine levels have been used in some cases where transient hypoglycaemia has proved difficult to document.

Insulin Assay

When a dog or cat presents in a hypoglycaemic crisis while management of the emergency is clearly important, obtaining samples to confirm a diagnosis is also crucial. If hypoglycaemia of <3 mmol/l is documented then the result should be verified and a serum sample should be taken for a contemporaneous insulin assay. The lower the blood glucose, the more essential this is. If a hypoglycaemic episode is only transient it is frustrating not to have collected the correct samples when the opportunity arose. While there have been attempts to identify a definitive ratio comparing insulin and glucose assays which would confirm a diagnosis of insulinoma this has not proved to be adequately specific or sensitive. The rule of thumb is that if the insulin level is inappropriate (often high but sometimes normal) for a given blood glucose then this would support the diagnosis.

Imaging

To confirm the diagnosis of insulinoma ideally the pancreatic mass should be identified and a search for metastatic disease carried out given reports that 40-50% of cases have visible metastatic disease at the time of surgery. Principal metastatic sites include the regional lymphatics (duodenal, mesenteric, hepatic and splenic lymph nodes), liver and peripancreatic omentum. Metastases to the lungs are rare but have been reported.

The most useful, readily available, non-invasive and relatively inexpensive imaging modality in practice is likely to be transabdominal ultrasonography but there are few useful reports documenting good sensitivity with this technique and many primary tumours are only small. All abdominal ultrasonography techniques are likely to be highly operator dependent, adversely affected by gas in the intestinal tract surrounding the area of interest (duodenum, stomach and colon) and challenging in large-breed dogs. Although the detection rate for insulinoma might be poor, it is a useful technique for ruling out other differentials for paraneoplastic hypoglycaemia and if no lesion is found and the index of suspicion for insulinoma is still high then surgery would be indicated. A report from 2005 compared ultrasonography with computed tomography (CT) imaging and found CT identified the primary insulinoma in 10/14 dogs while only 5/13 dogs had pancreatic lesions identified with ultrasonography. The ability to accurately identify lymph node metastasis was less convincing with many false-positive findings on CT. Limitations with CT in the human field include the failure of small tumours to adequately distort the contour of the pancreas and minimal difference in attenuation between lesions and normal pancreatic parenchyma on pre- and post- contrast scans.

It has been reported that in people preoperative localisation imaging studies with transabdominal ultrasonography, CT and angiography failed in 40-60% of patients. Endoscopic ultrasonography may be a developing technique in the human field with a report stating an overall sensitivity and accuracy of 93% for pancreatic neuroendocrine tumours, and perhaps this technique will be applicable to our patients in the future. While surgical debulking still remains the treatment of choice even in the face of metastatic disease, the case for accurate imaging assessments will always be debated.

Management: The Hypoglycaemic Crisis

While the emphasis has so far been on differentials and diagnostics, it goes without saying that many hypoglycaemic presentations require emergency management.

Dextrose

Appropriate supplementation to correct the hypoglycaemia is recommended, but beware of 'over treating' which can be counterproductive, stimulating further release of insulin resulting in rebound hypoglycaemia. In a genuine emergency approximately 1-2 ml/kg of 50% dextrose solution (diluted) can be given as a slow intravenous bolus and should achieve a rapid effect. A dextrose supplemented (2.5-10%) intravenous fluid infusion should continue but making sure to use a central line for concentrations >5% to avoid complications such as thrombophlebitis. Additional management will need to be introduced along with appropriate discussions with the owner once the patient is stable.

Glucagon

Where it has proved impossible to stabilise a hypoglycaemic patient with a dextrose bolus/ infusion then glucagon constant rate infusion (CRI) may be appropriate as a physiological and rational approach to management. The protocol described involves making a 1µg/ml solution (1 mg vial of glucagon in 1 litre of 0.9%NaCl) and suggests dose rates of 5 ng/kg/min to 10-13 ng/ kg/min. Close monitoring would be required on a case-by-case basis. This protocol is promoted as being less likely to induce swings of hypo-and hyperglycaemia, both of which could further compromise damaged brain tissue in dogs that have had seizures. While euglycaemia has been reported to occur within 1 hour, the CRI was continued but tapered over 4 days and effects have lasted up to 3 weeks.

Management of the Stable Patient

 Surgery: diagnostic and therapeutic for definitive treatment or debulking

 Dietary management: small frequent feeds of high fat, protein and complex carbohydrate diet to minimise postprandial hyperglycaemia

 Prednisolone: 0.2-0.5mg/kg q24h or q12h to induce peripheral insulin resistance, stimulate glycogenolysis and provide substrates for gluconeogenesis (cheap and often helpful)

 Diazoxide: inhibits release of insulin (expensive, not readily available and side effects)

 Octreotide (somatostatin analogue): variable reports regarding efficacy of treatment which is dependent on presence of appropriate somatostatin receptors on the tumour cells

 Streptozotocin: specific cytotoxic agent for beta cells and therefore diabetogenic as well as nephrotoxic and side effects include vomiting (current role unclear)

References

1.  Anderson M, Carpenter S, et al. Endoscopic ultrasound is highly accurate and directs management in patients with neuroendocrine tumours of the pancreas. The American Journal of Gastroenterology 2000; 95: 2271-2277.

2.  Feldman EC, Nelson RW. Beta cell neoplasia: Insulinoma. In: Feldman, EC; Nelson, RW. eds. Canine and feline endocrinology and reproduction (third edition). Philadelphia: WB Saunders, 2004; 616-644.

3.  Fischer JR, Smith SA, Harkin KR. Glucagon constant rate infusion: a novel strategy for the management of hyperinsulinaemic-hypoglycaemic crisis in the dog. Journal of the American Animal Hospital Association 2000; 36: 27-32.

4.  Moore AS, Nelson RW, et al. Streptozotocin for treatment of pancreatic islet cell tumours in dogs: 17 cases (1989-1999) Journal of the American Veterinary Medical Association 2002; 221: 811-818.

5.  Robben JH, Pollak YWEA, et al. Comparison of ultrasonography, computed tomography and single photon emission computed tomography for the detection and localisation of canine insulinoma. Journal of Veterinary Internal Medicine 2005; 19: 15-22.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Jane Eastwood, MA, VetMB, DVR, DSAM, MRVCS
The Royal Veterinary College
Hatfield, Hertfordshire, UK


MAIN : Endocrinology : Beta Cell Neoplasia
Powered By VIN
SAID=27