Link to Insulin Teaching Resources
Link to Problems for Discussion



  • Insulin Aspart Protamine, Recombinant/Insulin Aspart, Recombinant
  • Insulin Aspart, Recombinant
  • Insulin Degludec1
  • Insulin Degludec/Insulin Aspart, Recombinant
  • Insulin Detemir
  • Insulin Glargine, Recombinant
  • Insulin Glulisine
  • Insulin Human Isophane (NPH)
  • Insulin Human Isophane (NPH)/Insulin Human Regular
  • Insulin Human Regular2
  • Insulin Lispro Protamine, Recombinant/Insulin Lispro, Recombinant
  • Insulin Lispro, Recombinant
1Also available in a 200 Unit/mL concentration (rather than the usual 100 U/mL)
2Also available in a 500 Unit/mL concentration (rather than the usual 100 U/mL)


As expected, the toxicity of insulin in overdose is primarily due to hypoglycaemia, although hypokalaemia may also cause problems. The duration of the hypoglycaemic effect depends on the type of insulin injected (duration of action), the amount and age, diabetes (insulin resistance) and other factors that may increase or decrease the patient`s sensitivity to insulin. Mortality in attempted suicidal overdose with insulin is 25%. Death has occurred after as little as 20 units but doses of 400 to 900 units or more are more common in fatal cases. Irreversible neurological injury occurs when glycogen stores are depleted since the brain is totally dependent on glucose metabolism. It is the duration of hypoglycaemia in the presence of signs or symptoms of neurological compromise that determines post hypoglycaemic encephalopathy, rather than the quantity of insulin injected. The period from injection of an overdose of insulin to irreversible brain damage is frequently about 7 hours (about the time glycogen stores are completely). Treatment is with 50% glucose IV, food (and lots of it) and close observation.


Insulin binds to insulin receptors and activates the glucose transporter, moving glucose into cells. It has numerous other metabolic effects which are of little relevance to in overdose; insulin may cause more effect in non-diabetics, who do not have insulin antibodies or insulin resistance (Stapczynski et al., 1984).



Insulin is not bioavailable if taken orally. Absorption is very rapid from IM administration, onset of effect is within minutes. Absorption after subcutaneous injection depends on volume of injection, smoking, exercise, site, and external temperature, as well as on the type of insulin. The lag time between injection and first appearance of insulin is about 15 min for soluble human insulin; the absorption half-life is about 90 min. Onset of action for regular insulin begins at about 30 min with maximum concentration reached about 1–2 h after subcutaneous injections. The time to peak hypoglycaemic activity is 2–5 hours; this varies considerably both within and between individuals. The duration of action is 6–8 hours. Subcutaneous administration is the usual route of administration in therapeutic use. The onset and duration of effect depend on the formulation (see pharmacokinetics).

The onset in overdose may be earlier than that expected in therapeutic use. However, in most insulin overdoses the duration of absorption is longer than that observed with the same preparation in therapeutic doses. For example, overdoses of NPH insulin (normal duration 18–24 hours) have required glucose supplementation for 6 days (Samuels & Eckel, 1989).


Insulins are 90 to 99% protein bound and have a volume of distribution of about 0.2 L/kg.

Metabolism - Elimination

The major difference between insulins is the duration of action, which is related to their absorption from the site where they are administered.

The elimination half-life of the insulin that reaches plasma is much shorter than the duration of hypoglycaemic action at about 4-6 min but the apparent half-life after subcutaneous injection is about 2 hours because of continuing absorption. Insulin is destroyed in the kidneys and liver.

The half-life of insulin is prolonged with renal impairment.

Pharmacokinetic parameters of importance to insulin overdose
(Figures are for therapeutic use (information from Tibaldi, 2012)

*Peak concentration may occur much later and duration may be much longer in overdose



The onset of major clinical effects from hypoglycaemia may be very rapid and with little warning. Insulin will cause hypoglycaemia in virtually all non-diabetics. Hypoglycaemia may develop within 30 minutes but may be delayed much later than predicted for the conventional duration of action of various insulin preparations. There is little correlation between insulin dose and severity of hypoglycaemia. This is often because of the marked change in absorption characteristics when large doses of insulin are injected in one place. The period from injection of an overdose of insulin to irreversible brain damage is frequently about 7 hours (about the time glycogen stores are completely).

Late onset of clinical symptoms should be uncommon, the exception being situations where large amounts of glucose have been given prior to the onset of symptoms and this has masked the presentation for some time. Glucose supplementation should not be ceased abruptly and patients should be closely monitored for several hours after it is discontinued. Clinical effects are predominantly due CNS effects of
hypoglycaemia (confusion, coma, seizures, cerebral oedema) and to the effects of reactive sympathetic stimulation (tachycardia, sweating, tremor, ischaemic chest pain).

Other effects

Aspiration pneumonia or pulmonary oedema is sometimes seen particularly after seizures or prolonged hypoglycaemia. Hypokalaemia, and other electrolyte imbalances, often occur during prolonged treatment with glucose.


Blood glucose

Conversion factor
  • mg/dL x 0.555 = mmol/L
  • mmol/L x 18 = mg/dL
Normal range 4.2–6.4 mmol/L (75–115 mg/dL)

Blood glucose should be measured at least hourly for at least 6 to 8 hours. Subsequent frequency will depend on the clinical situation and previous results. If glucose treatment is required, then measurements should be made for 6 to 8 hours after glucose has been ceased.


Renal function should be measured as insulin is partly renally eliminated. Poor renal function will lead to a longer time course and potentially more severe hypoglycaemia. Hypokalaemia may occur if large amounts of glucose are required. Insulin and glucose move potassium from the extracellular to the intracellular compartment. Hyponatraemia may occur from excessive water given with glucose.

Blood concentrations

Blood concentrations of insulin are unhelpful in acute management but may be able to differentiate hypoglycaemia from overdose from therapeutic misadventures. Insulin, C-peptide and proinsulin can help differentiate between exogenous insulin overdose and insulinoma.
C-peptide concentrations can help decide when to stop glucose infusions in non-diabetics with insulin overdose by indicating when endogenous insulin release has commenced (which means exogenous insulin has gone).


The ECG should be monitored if significant hypokalaemia occurs.


A rapid response to glucose in a patient who has coma and/or seizures strongly suggests hypoglycaemia. There are a number of other drugs that may cause hypoglycaemia in overdose other than insulin and sulphonylureas. Beta-blockers, salicylates and alcohol would be the most important drugs to consider but should be readily differentiated on clinical grounds. Factitious hypoglycaemia (overdose with sulphonylureas or insulin overdose) and insulinoma can be distinguished by a combination of insulin, C-peptide and drug concentrations (Lebowitz & Blumenthal, 1993).


The major difference between insulins relates to their onset and duration of action (see pharmacokinetic properties). Mortality in attempted suicidal overdose with insulin is 25%. Death has occurred after as little as 20 units but doses of 400 to 900 units or more are more common in fatal cases.



IV access should be secured as soon as possible. If there is any doubt as to whether the patient is thiamine deficient (e.g. alcohol abuse, anorexia nervosa), then thiamine should be given IMI. If prolonged use of very hypertonic solutions (i.e. 25 or 50% glucose) is likely to be required, then a central line should be inserted.


Intravenous glucose (50%) should be given to all symptomatic patients. A prolonged infusion may be necessary. Lower concentrations of glucose have very low caloric value and will be unable to maintain blood sugar without requiring massive water overload. If patients are able to eat, they should be encouraged to eat small but very frequent carbohydrate-based meals (see caloric content of IV glucose and some foods).
The blood glucose should be maintained above 3.5 mmol/L (60 mg/dL).

GI Decontamination

If the insulin has been swallowed, decontamination is not necessary.

Surgical excision

In massive insulin injection, local excision of insulin injection site has been suggested as a means to to reduce insulin exposure, especially in long-acting insulin overdose. Excision performed down to the muscle wall is then generally recommended.

For this technique to be effective, large doses of long-acting insulin must have been injected subcutaneously into a single site. In that circumstance, insulin has been surgically removed leading to a shorter than expected duration of hypoglycaemia (Campbell & Ratcliffe 1982).

However, with current preparations of insulin, one no longer sees such massive injections into the same site and thus we do not recommend local excision of insulin injection sites.


Glucagon can be used but is not recommended in large overdoses. It relies on hepatic glycogen stores for its effect and in severe overdoses these are likely to be exhausted. Corticosteroids are ineffective for similar reasons. Glucagon also stimulates insulin release from the pancreas (in patients with functioning islet cells) and thus may potentially worsen the situation.

Treatment of specific complications

These should be treated with IV glucose. Late onset or prolonged seizures may be due to CNS damage. If there is no hypoglycaemia, then diazepam and phenytoin would be appropriate. Phenytoin also inhibits the secretion of insulin from the pancreas in high doses! (Malherbe
et al, 1972).

Cerebral oedema & severe brain injury
If there is no other drug ingestion, the patient who remains comatose despite normal blood glucose has generally sustained a significant neurological insult. Raised intracranial pressure often occurs in this situation. The treatment is similar to raised intracranial pressure from any brain injury and includes hyperventilation, raising the head of the bed, blood pressure control, mannitol and fluid restriction. In this context, it is important that the treatment of hypoglycaemia is not exacerbating the neurological injury. So the highest concentrations of glucose (50%) only should be used, and if diazoxide or sodium bicarbonate is given this should be done cautiously to avoid fluid overload.

A central venous line is essential to administer glucose and monitor hydration.

Elimination enhancement

Insulin has a very short intravascular half-life and there is no benefit from repeated doses of activated charcoal, haemodialysis, haemoperfusion or urinary pH changes.


Long term sequelae commonly occur if there has been a prolonged period of hypoglycaemia. Irreversible neurological injury occurs when glycogen stores are depleted since the brain is totally dependent on glucose metabolism. It is the duration of hypoglycaemia in the presence of signs or symptoms of neurological compromise that determines post hypoglycaemic encephalopathy, rather than the quantity of insulin injected. Patients should be routinely followed up. If there are physical signs or the patient complains of neurological symptoms, neuropsychological evaluation should be performed.

Almost any type of neurological damage may occur, although lesions have typically been reported in "watershed" areas of the brain (Kaiser et al., 1981).


Arem R, Zoghbi W. Insulin overdose in eight patients: insulin pharmacokinetics and review of the literature. Medicine 1985; 64(5):323-32.
Campbell IW, Ratcliffe JG. Suicidal insulin overdose managed by excision of insulin injection site. Br Med J 1982; 285(6339):408-9.
Hardman JG, Gilman AG, Limbird LE. Gooodman and Gilman's The pharmacological basis of therapeutics 9th ed. 1996. McGraw Hill, New York.
Jefferys DB, Volans GN. Self-poisoning in diabetic patients. Hum Toxicol 1983; 2:345-348.
Kaiser MC, Pettersson H, Harwood-Nash DC, Fitz CR, Chuang S. Case report. Computed tomography of the brain in severe hypoglycaemia. J Computer Assisted Tomography 1981; 5(5):757-9.
Lebowitz MR, Blumenthal SA. The molar ratio of insulin to C-peptide. An aid to the diagnosis of hypoglycaemia due to surreptitious (or inadvertent) insulin administration. Arch Int Med 1993; 153: 650-5.
Malherbe C, Burril KC, Levin SR, Karam JH, Forsham PH. Effect of diphenylhydantoin on insulin secretion in man. N Engl J Med 1972; 286:339.
Moore DF, Wood DF, Volans GN. Features, prevention and management of acute overdose due to antidiabetic drugs. Drug Safety 1993; 9: 218-229.
Samuels MH, Eckel RH. Massive insulin overdose: detailed studies of free insulin concentrations and glucose requirements. J Toxicol Clin Toxicol 1989;27: 157-168.
Scheen AJ, Lefebvre PJ. Antihyperglycaemic agents. Drug interactions of clinical importance. Drug Safety 1995; 12(1):32-45.
Seger D. Toxic emergencies of endocrine and metabolic therapeutic agents. J Emerg Med 1988; 6(6):527-37.
Stapczynski JS, Haskell RJ. Duration of hypoglycaemia and need for intravenous glucose following intentional overdoses of insulin. Ann Emerg Med 1984; 13(7):505-11.
Tibaldi JM. Evolution of insulin development: focus on key parameters. Adv Ther. 2012 Jul;29(7):590–619.