Not Found

Find information on medical topics, symptoms, drugs, procedures, news and more, written for the health care professional.

* This is the Professional Version. *


By Erika Brutsaert, MD, Assistant Professor;Attending Physician, Albert Einstein College of Medicine;Montefiore Medical Center

Click here for
Patient Education

Hypoglycemia unrelated to exogenous insulin therapy is an uncommon clinical syndrome characterized by low plasma glucose level, symptomatic sympathetic nervous system stimulation, and CNS dysfunction. Many drugs and disorders cause it. Diagnosis requires blood tests done at the time of symptoms or during a 72-h fast. Treatment is provision of glucose combined with treatment of the underlying disorder.

Most commonly, symptomatic hypoglycemia is a complication of drug treatment of diabetes mellitus. Oral antihyperglycemics or insulin may be involved.

Symptomatic hypoglycemia unrelated to treatment of diabetes mellitus is relatively rare, in part because the body has extensive counter-regulatory mechanisms to compensate for low blood glucose levels. Glucagon and epinephrine levels surge in response to acute hypoglycemia and appear to be the first line of defense. Cortisol and growth hormone levels also increase acutely and are important in the recovery from prolonged hypoglycemia. The threshold for release of these hormones is usually above that for hypoglycemic symptoms.


Causes of physiologic hypoglycemia can be classified as

  • Reactive (postprandial) or fasting

  • Insulin-mediated or non–insulin-mediated

  • Drug-induced or non–drug-induced

Insulin-mediated causes include exogenous administration of insulin or an insulin secretagogue and insulin-secreting tumors (insulinomas).

A helpful practical classification is based on clinical status: whether hypoglycemia occurs in patients who appear healthy or ill. Within these categories, causes of hypoglycemia can be divided into drug-induced and other causes.

Non–islet cell tumor hypoglycemia (NICTH) is a rare cause of hypoglycemia in which excess insulin growth factor 2 (IGF-2) produced by the tumor is the cause of hypoglycemia.

Pseudohypoglycemia occurs when processing of blood specimens in untreated test tubes is delayed and cells, such as RBCs and leukocytes (especially if increased, as in leukemia or polycythemia), consume glucose. Factitious hypoglycemia is true hypoglycemia induced by nontherapeutic administration of sulfonylureas or insulin.

Symptoms and Signs

The surge in autonomic activity in response to low plasma glucose causes sweating, nausea, warmth, anxiety, tremulousness, palpitations, and possibly hunger and paresthesias. Insufficient glucose supply to the brain causes headache, blurred or double vision, confusion, difficulty speaking, seizures, and coma.

In controlled settings, autonomic symptoms begin at or beneath a plasma glucose level of about 60 mg/dL (3.3 mmol/L), whereas CNS symptoms occur at or below a glucose level of about 50 mg/dL (2.8 mmol/L). However, symptoms suggestive of hypoglycemia are far more common than the condition itself. Most people with glucose levels at these thresholds have no symptoms, and most people with symptoms suggestive of hypoglycemia have normal glucose concentrations.


  • Blood glucose level correlated with clinical findings

  • Response to dextrose (or other sugar) administration

  • Sometimes 48- or 72-h fast

  • Sometimes insulin, C-peptide, and proinsulin levels

In principle, diagnosis requires verification that a low plasma glucose level (< 50 mg/dL [< 2.8 mmol/L]) exists at the time hypoglycemic symptoms occur and that the symptoms are responsive to dextrose administration. If a practitioner is present when symptoms occur, blood should be sent for glucose testing. If glucose is normal, hypoglycemia is ruled out and no further testing is needed. If glucose is abnormally low, serum insulin, C-peptide, and proinsulin measured from the same tube can distinguish insulin-mediated from non–insulin-mediated and factitious from physiologic hypoglycemia and can obviate the need for further testing.

In practice, however, it is unusual that practitioners are present when patients experience symptoms suggestive of hypoglycemia. Home glucose meters are unreliable for quantifying hypoglycemia, and there are no clear glycosylated Hb (HbA1c) thresholds that distinguish long-term hypoglycemia from normoglycemia. So the need for more extensive diagnostic testing is based on the probability that an underlying disorder that could cause hypoglycemia exists given a patient’s clinical appearance and coexisting illnesses.

A 72-h fast done in a controlled setting is the standard for diagnosis. However, in almost all patients with a hypoglycemic disorder, a 48-h fast is adequate to detect hypoglycemia and a full 72-h fast may not be necessary. Patients drink only noncaloric, noncaffeinated beverages, and plasma glucose is measured at baseline, whenever symptoms occur, and every 4 to 6 h or every 1 to 2 h if glucose falls below 60 mg/dL (3.3 mmol/L). Serum insulin, C-peptide, and proinsulin should be measured at times of hypoglycemia to distinguish endogenous from exogenous (factitious) hypoglycemia. The fast is terminated at 72 h if the patient has experienced no symptoms and glucose remains normal, sooner if glucose decreases to 45 mg/dL ( 2.5 mmol/L) in the presence of hypoglycemic symptoms.

End-of-fast measurements include beta-hydroxybutyrate (which should be low in insulinoma), serum sulfonylurea to detect drug-induced hypoglycemia, and plasma glucose after IV glucagon injection to detect an increase characteristic of insulinoma. Sensitivity, specificity, and predictive values for detecting hypoglycemia by this protocol have not been reported.

There is no definitive lower limit of glucose that unequivocally defines pathologic hypoglycemia during a monitored fast. Normal women tend to have lower fasting glucose levels than men and may have glucose levels as low as 30 mg/dL (1.7 mmol/L) without symptoms. If symptomatic hypoglycemia has not occurred by 48 to 72 h, the patient should exercise vigorously for about 30 min. If hypoglycemia still does not occur, insulinoma is essentially excluded and further testing is generally not indicated.


  • Oral sugar or IV dextrose

  • Sometimes parenteral glucagon

Immediate treatment of hypoglycemia involves provision of glucose. Patients able to eat or drink can drink juices, sucrose water, or glucose solutions; eat candy or other foods; or chew on glucose tablets when symptoms occur. Infants and younger children may be given 10% dextrose solution 2 to 5 mL/kg IV bolus. Adults and older children unable to eat or drink can be given glucagon 0.5 (< 20 kg) or 1 mg ( 20 kg) sc or IM or 50% dextrose 50 to 100 mL IV bolus, with or without a continuous infusion of 5 to 10% dextrose solution sufficient to resolve symptoms. The efficacy of glucagon depends on the size of hepatic glycogen stores; glucagon has little effect on plasma glucose in patients who have been fasting or who are hypoglycemic for long periods.

Underlying disorders causing hypoglycemia must also be treated. Islet cell and non–islet cell tumors must first be localized, then removed by enucleation or partial pancreatectomy; about 6% recur within 10 yr. Diazoxide and octreotide can be used to control symptoms while the patient is awaiting surgery or when a patient refuses or is not a candidate for a procedure.

Islet cell hypertrophy is most often a diagnosis of exclusion after an islet cell tumor is sought but not identified.

Drugs that cause hypoglycemia, including alcohol, must be stopped.

Treatment of hereditary and endocrine disorders, liver failure, renal failure, heart failure, sepsis, and shock are described elsewhere.

Key Points

  • Hypoglycemia is low plasma glucose level (< 50 mg/dL [< 2.8 mmol/L]) plus simultaneous hypoglycemic symptoms that reverse with dextrose administration.

  • Most hypoglycemia is caused by drugs used to treat diabetes mellitus (including surreptitious use); insulin-secreting tumors are rare causes.

  • If etiology is unclear, do a 48- or 72-h fast with measurement of plasma glucose at regular intervals and whenever symptoms occur.

  • Measure serum insulin, C-peptide, and proinsulin at times of hypoglycemia to distinguish endogenous from exogenous (factitious) hypoglycemia.

* This is the Professional Version. *