LabMed

Hypoglycemia

At a Glance

Genuine hypoglycemia in adults is a diagnosis that requires establishing Whipple's Triad:

  • clinical symptoms consistent with hypoglycemia (i.e., anxiety, palpitations, tremor)

  • low blood glucose value (typically <55 mg/dL)

  • relief of symptoms with administration of glucose

In particular, it is important not to make this diagnosis based solely on a low blood glucose value. Some patients can tolerate relatively low glucose values without symptoms. And some glucose values can be erroneous, either because methods with poor precision at low ranges are used (some glucose meters) or because of specimen integrity issues (see pseudohypoglycemia).

What Tests Should I Request to Confirm My Clinical Dx? In addition, what follow-up tests might be useful?

In patients with symptoms whose blood glucose is genuinely low, there are two major types of hypoglycemia: those with inappropriately high insulin levels (hyperinsulinemic hypoglcyemia) and those with appropriately low insulin levels (hypoinsulinemic hypoglycemia).

No matter what the cause of hypoglycemia is, a standard panel of laboratory tests can be done to make a definitive diagnosis as to etiology. If a patient has an episode during your visit, you can document the symptoms, draw the necessary blood samples, and administer glucose to document the resolution of symptoms (Whipple's Triad).

More likely, the patient will not exhibit symptoms during the visit. Once you can piece together the scenarios under which symptoms usually occur (e.g., after prolonged fasting versus within a few hours following a meal), you can duplicate the conditions and be ready to have the samples drawn and have glucose administered to document resolution of the symptoms. Typically, this involves a 72-hour fast under controlled conditions, enabling administration of glucagon intravenously at the conclusion of the study.

The entire battery of tests listed in textbooks and online resources includes glucose, insulin, hypoglycemic agent screen, C-peptide, beta-hydroxybutyrate, and proinsulin. However, it should be noted that all of these tests are not needed to make a diagnosis and the tests (other than glucose) should be run only on the sample with low glucose. By following a stepwise, logical progression, you can limit tests (and the associated expense and potential confusion) to those relevant to each patient:

It is critical to keep in mind that insulin levels measured on other samples can confound the diagnosis, as samples drawn after glucose has been administered to relieve hypoglycemia will likely have measurable insulin levels that are totally appropriate.

What Lab Results Are Absolutely Confirmatory?

For the remainder of this discussion, let's assume you have a properly drawn sample with a low glucose value (<55 mg/dl) from a patient in whom you have demonstrated Whipple's Triad. The next logical step is to determine the insulin level on that sample.

If the insulin level is inappropriately high (>0.3 uU/mL), you can infer that the patient has hyperinsulinemic hypoglycemia. Since C-peptide is produced when insulin is synthesized in vivo but is not found in commercial insulin preparations, a low C-peptide level (<0.2 nmol/L) enables you to infer that the patient has taken exogenous insulin (administered as a drug, as would occur if a patient were given insulin or if a patient took insulin surreptitiously). In contrast, high levels of C-peptide indicate an endogenous source (as with patients on hypoglycemic agents or patients with insulin-producing tumors).

In the case of endogenous insulin production, it is worth noting that hypoglycemic agents are a much more common explanation than insulinomas, thus, emphasizing the importance of obtaining a comprehensive hypoglycemic agent screen in patients with inappropriately high insulin and C-peptide levels.

If, on the sample with low glucose, the insulin concentration is appropriately low (<0.3 uU/mL), you are dealing with hypoinsulinemic hypoglycemia. The two possible explanations (insulin-like agents, such as IGF-II, versus severe intrinsic liver disease) should be relatively easy to distinguish clinically. As further supporting evidence, however, you can order beta-hydroxybutyrate levels on the sample with low glucose and/or evaluate the glucose levels obtained following glucagon administration.

In the case of severe liver disease (i.e., a liver with no glycogen stored), glucose levels will not rise in response to glucagon. In addition, because of the absence of insulin effects, beta-hydroxybutyrate levels will be high as a result of fat metabolism induced by the fast. In contrast, if an insulin-like protein is responsible, glucose levels will respond to glucagon and the beta-hydroxybutyrate levels will be low.

It is worth calling attention to the special case of ethanol, which would be categorized as "severe liver disease." Although ethanol can inhibit gluconeogenesis on its own, the propensity to cause hypoglycemia is more often seen in patients with chronic alcoholism, in whom malnutrition (low glycogen stores) is also present.

What Tests Should I Request to Confirm My Clinical Dx? In addition, what follow-up tests might be useful?

Other causes for genuine hypoglycemia, although cited in comprehensive reviews, are exceedingly rare.

In the case of hyperinsulinemic hypoglycemia, it is possible for patients to make autoantibodies to insulin. (These are distinct from the antibodies made to exogenous insulin by patients with diabetes.) In the presence of these antibodies, insulin secreted in response to hyperglycemia is thought to be bound by these antibodies for variable amounts of time, sometimes getting released when the glucose level has already returned to normal, resulting in hypoglycemia and hyperinsulinemia (with high C-peptide). If this explanation is suspected, any sample from the patient (i.e., it need not be a sample with low glucose) can be tested for the presence of these antibodies at any time.

In the case of hypoinsulinemic hypoglycemia, deficiencies of cortisol and/or growth hormone are sometimes cited as potential explanations. These two hormones do serve as defenses after prolonged hypoglycemia (hours), following the primary defenses of decreased insulin secretion, increased glucagon secretion, and increased epinephrine secretion. It should be noted, however, that it is exceptionally rare for deficiencies in cortisol (e.g., Addison's disease) or growth hormone to cause hypoglycemia.

You must be a registered member of Infectious Disease Advisor to post a comment.

Sign Up for Free e-newsletters