Acute exenatide (Byetta®) poisoning was not associated with significant hypoglycemia

Acute exenatide (Byetta®) poisoning was not associated with significant hypoglycemia

To the Editor:

Exenatide, a self-injectable pen device (1), is an incretin-mimetic derived from the venom of the Gila monster (Heloderma suspectum). It works by mimicking the action of glucagon-like peptide 1 (GLP1) and is the first of a new class of anti-diabetic medications. We report the first probable case of an acute overdose with exenatide that was not associated with hypoglycemia. A 40-year-old morbidly obese woman presented to the emergency department claiming to have injected 10 mcg of exenatide nine times in an attempted suicide. This was in addition to routine administration of her morning medications. She denied having any breakfast and stated feelings of hopelessness and helplessness.

Her past medical history includes Diabetes Mellitus Type II, depression, hypertension, asthma, and hyperlipidemia. Her prior medications included furosemide 40 mg daily, irbesartan 300 mg daily, pioglitazone 25 mg daily, metformin 1000 mg twice daily, glimepiride 4 mg daily, escitalopram 20 mg daily, alprazolam 0.5 mg daily, atorvastatin 20 mg daily, fenofibrate 145 mg daily, pantoprazole 40 mg twice daily, subcutaneously administered exenatide 10 mcg daily, lantus 30 U daily, and inhaled fluticasone proprionate/salmeterol xinofoate diskus 500/50 mcg, one puff twice daily.

At triage, the patient had symptoms of dizziness, weakness, nausea, and one episode of vomiting. Initial vital signs were normal and stable, and fasting blood glucose ranged from 84 to 109 mg/dl. The results of her physical examination were unremarkable except for a hematoma with eight small incisions on her left thigh. Laboratory values were within normal limits.

Initial management recommendations were to monitor blood glucose levels hourly and provide supportive therapy with dextrose as needed for hypoglycemia. The only therapeutic intervention required was metoclopramide 10 mg intravenously. Postprandial blood glucose levels rose to 162 mg/dl. The patient was admitted with a diagnosis of suicidal attempt and bipolar disorder, and never became hypoglycemic.

Three cases of inadvertent delivery of 10 times the normal dose of exenatide were reported during a phase III randomized open-label study (2). All patients experienced severe nausea and vomiting with one categorized as severely hypoglycemic. Severe hypoglycemia was noted if a third party intervention was required as prophylaxis. In this case, 5% dextrose in water with normal saline was given in response to the patient's severe vomiting. Blood glucose levels were not provided (personal communication with Dr. Dennis Kim, Medical Director, June, 22, 2006).

The incidence of hypoglycemia in phase III clinical trials using therapeutic doses of exenatide was 5% to 36% (1,3–7). Therefore, one would expect that toxic doses should exacerbate the normal pharmacological actions seen at therapeutic doses. Thus, why did our patient remain euglycemic?

The lack of hypoglycemia observed in our case may be due to the patient's failure to successfully self-administer the exenatide. An initial six-step process is needed to prepare the pen, and then a subsequent 10-step process is needed to administer each dose (8). Another explanation is that the patient prevaricated.

Alternatively, the lack of feeding prior to exenatide may account for this patient's euglycemic state. GLP 1, a 39-amino acid peptide, mimics the enteric hormone secreted from the intestine (L-cells) in response to nutrient ingestion, stimulating glucose-dependent insulin release and suppression of glucagon secretion (1,5). Our patient purported not to have consumed any food before taking exenatide, thus omitting any stimulant to insulin release and any risk of hypoglycemia.

Exenatide has been associated with inducing delayed gastric emptying of various pharmacological agents (1,5). Thus, exenatide may have reduced bioavailability of the patient's oral antidiabetic medications. Delayed gastric emptying induced by exenatide in the overdose setting has not been studied. However, therapeutic doses of exenatide were administered in combination with acetaminophen and lovastatin, delayed gastric emptying, and reduced the bioavailability of both of these agents (6). Data of reduced bioavailability with oral antidiabetics and how this alters glycemic control are not well understood.

The symptoms of toxicity were temporally related to the time of self-injection of exenatide. Exenatide reaches peak concentrations in 2.1 hours and is eliminated via glomerular filtration, with a terminal half-life of 2.4 hours (1). No re-challenge was conducted, as it would be unethical to administer a known toxic dose. Furthermore, it is unlikely that any other medications or underlying co-morbid conditions could have attributed to this patient's presentation other than the self-injection of exenatide.

Based on Naranjo algorithm (9), this adverse event is classified as probable. This event resulted in a lengthy hospital stay probably related to the patient's psychiatric condition.

This is the first post-marketing case of acute overdose with exenatide that was associated with a euglycemic state and minimal toxicity.