At a Glance

Rebound hypoglycemia was first described in 1959 by Dr. Michael Somogyi, which is how it came to be called the Somogyi phenomenon. His hypothesis was that, in patients with diabetes, hyperglycemia sometimes occurs as a result of an exaggerated hormonal response to hypoglycemia. In particular, the hypothesis sought to explain the observation that hospitalized patients with diabetes often had markedly elevated fasting blood glucose in the morning, despite having been given ever-increasing doses of insulin.

According to the hypothesis, these patients received too much, rather than too little, insulin, as a result of which they became hypoglycemic late at night or early in the morning. In response to the hypoglycemia, these patients produced inappropriately high levels of counter-regulatory hormones, such as glucagon and cortisol, resulting in high morning glucose levels. By giving these patients more insulin, one actually made the problem worse.

Over the past few decades, this hypothesis has been largely discredited. A paper published in 1984 had a particularly interesting title, “The Somogyi Phenomenon: Sacred Cow or Bull?” reflecting the author’s opinion that, despite limited evidence to support it, it was among the most widespread concepts in clinical medicine.

Continue Reading

The most persuasive evidence disproving the existence of the Somogyi phenomenon was provided more recently by continuous glucose monitoring. Although mild nocturnal hypoglycemia has been documented in some patients with diabetes, no significant increases in counter-regulatory hormones have been documented, nor have unexpectedly high levels of fasting glucose been observed. In retrospect, Dr. Somogyi’s observations and his successful treatment of some patients by lowering their doses of insulin have been explained by the relatively poor methods of monitoring glucose at that time (urine glucose) and to falling levels of insulin in the morning hours.

The Somogyi phenomenon is sometimes confused with the dawn phenomenon, a similar, but distinct, entity. The dawn phenomenon is invoked to explain what appear to be inappropriately high morning fasting glucose levels in patients with diabetes that are unrelated to antecedent hypoglycemia. In the case of the dawn phenomenon, normal early morning increases in cortisol and growth hormone are thought to be responsible for the observed mild increases in fasting glucose. The distinction between the Somogyi phenomenon and the dawn phenomenon may seem subtle, but it is one worth making in that prescribing more insulin to a patient who is already experiencing hypoglycemia is dangerous.

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

To distinguish among the Somogyi phenomenon, the dawn phenomenon, and inadequate insulin dosing (amount and timing), frequent blood glucose monitoring during nighttime hours is needed. In the absence of documented frank hypoglycemia (<55 mg/dL), the Somogyi phenomenon cannot be the explanation. Whether one calls it the dawn phenomenon or not, it should be possible to use the observed glucose levels during the night and first thing in the morning to tailor insulin dosing to provide better glucose control throughout these hours.