Defining Hypoglycaemia

When discussing hypoglycaemia in general, but particularly when discussing study results, much confusion arises from the fact that there is no uniform definition of hypoglycaemia. Most people would agree that symptoms and a low glucose level are part of the definition. However, the symptoms associated with hypoglycaemia are highly varied and rather unspecific and sometimes symptoms can be remarkably absent in those with the more severe forms of hypoglycaemia. Likewise, the glucose cut-off level that separates hypoglycaemia from normoglycaemia is subject to discussion, and may indeed differ between individuals and within individuals over time, depending on factors such as level of glycemic control, age and duration of diabetes.

Figure 1. The surgeon Allen O. Whipple, who was the first to systematically define hypoglycaemia.
Figure 1. The surgeon Allen O. Whipple, who was the first to systematically define hypoglycaemia.
The first to define hypoglycaemia was the pancreatic surgeon Allen Whipple (figure 1). As a pancreatic surgeon, he probably felt the need for a clear definition of hypoglycaemia when performing operations for suspected insulinoma. According to Whipple's triad an event qualifies as a hypoglycaemic event when:

  1. There is a low blood glucose level
  2. There are symptoms consistent with hypoglycaemia
  3. Symptoms rapidly disappear upon the administration of glucose

However, the first two parts of this definition are not as clear as one might wish.

Glucose cut-off levels

Most experts will agree that hypoglycaemia is not present when glucose is 4 mmol/L or above. However, it is more difficult to establish when hypoglycaemia is definitely present; again, most experts will agree that hypoglycaemia is definitely present at glucose levels of 2.8 mmol/L or lower. This leaves us with a grey zone where arguments can be made for or against the presence of hypoglycaemia [1][2]. While the exact cut-off may seem trivial for the individual patient and care-giver, who will factor in other arguments such as the presence and burden of symptoms, the exact glucose cut-off level has a profound influence on the overall epidemiology of hypoglycaemia, with higher cut-off levels increasing, and lower cut-off levels reducing the apparent frequency of hypoglycaemia. The American Diabetes Association (ADA) advocates a plasma glucose-cut-off level of 3.9 mmol/L. This recommendation is based on the fact that in those without diabetes, the levels of counterregulatory hormone levels such as glucagon that increase glucose, are starting to go up when glucose falls below this level. Shortly, it is argued that since the body tries to raise the glucose level, the glucose level must be too low. A further argument for this relatively high glucose cut-off level is that those with diabetes who use insulin or insulin secretagogues are unable to decrease their plasma insulin, which in healthy subjects would occur at a glucose-level of around 4.5 mmol/L. However, many physicians and societies around the world feel that the glucose cut-off level chosen by the ADA is too high. Population studies have shown that in perfectly healthy individuals, glucose levels as low as 2.8 mmol/L can be found without any symptoms or adverse consequences[3]. A risk of setting too high a threshold for hypoglycaemia is that it will hamper attempts to lower average glycaemia in those with diabetes. Interestingly, several studies have shown that the thresholds at which an increase in counterregulatory hormones occurs may in fact vary between individuals. Those with longstanding poor glycemic control tend to have a counterregulatory response at higher glucose levels than normal, while those with longstanding tight control tend to have this response at lower glucose levels than normal[4]. While the various societies and those designing clinical studies may chose their favoured glucose cut-off level for a uniform definition of hypoglycaemia (varying between 3.0 and 3.9 mmol/L), in clincal practice it is probably more useful to define hypoglycaemia at the individual patient level. Thus, in a young person with diabetes of short duration, who has an intact counterregulatory system and who suffers only rarely from hypoglycaemia, a low cut-off level of 3.1 mmol/L can be quite acceptable, whereas in a frail old patient with long duration of diabetes and poor counterregulatory and poor autonomic function the risk of hypoglycaemia-associated falls may lead to a more cautious cut-off level of 3.9 or even 4.5 mmol/L.

Symptoms consistent with hypoglycaemia

Hypoglycaemia can present with a wide range of symptoms. However, while for the individual patient symptoms can be remarkably consistent, symptoms of hypoglycaemia are rather unspecific. Thus it is not rare for a patient to identify symptoms as 'hypoglycaemic' only to find (when checking the glucose level) that he/she is in fact hyperglycaemic. This is not so surprising. Hypoglycaemic symptoms broadly fall in two categories: autonomic symptoms that are mediated by the autonomic response and neuroglycopenic symptoms /cerebral dysfunction that results from glucose deficiency in the brain. However, hypoglycaemia is not the only trigger for sympathetic neural activation and 'typical' neuroglycopenic symptoms such as fatigue or lack of concentration likewise have many other causes. Moreover, with long-standing diabetes and repeated hypoglycaemia, patients may develop impaired counterregulation and hypoglycaemia unawareness. Studies have clearly shown that patients with type 1 diabetes loose the glucagon response to hypoglycaemia after a duration of diabetes of about 5 years, followed by attenuation of the adrenaline response after about 15 years[5]. The secretion of cathecholamines may still occur, although usually at lower blood glucose levels.Thus, adrenergic symptoms will diminish with longer standing type 1 diabetes. In patients with insulin deficient type 2 diabetes epinephrine responses have been found intact. However, recent antecedent hypoglycaemia was shown to shift the glycaemic treshold for epinephrine release to lower plasma glucose levels[6]. Aging itself is another risk factor for loss of autonomic function and diminished counterregulation.

Thus, paradoxically, those with the highest risk of (severe) hypoglycaemia may develop hypoglycaemia without any symptoms at all.


  1. ^ Cryer PE: Preventing hypoglycaemia: what is the appropriate glucose alert value? Diabetologia. 2009 Jan;52(1):35-7.

  2. ^ Frier BM: Defining hypoglycaemia: what level has clinical relevance? Diabetologia. 2009 Jan;52(1):31-4.

  3. ^ Merimee TJ, Fineberg SE. Homeostasis during fasting. II. Hormone substrate differences between men and women. J Clin Endocrinol Metab 1973;37:698-702

  4. ^ Boyle PJ et al.: Plasma glucose concentrations at the onset of hypoglycemic symptoms in patients with poorly controlled diabetes and in nondiabetics. N Engl J Med. 1988.

  5. ^ Bolli G et al.: Abnormal glucose counterregulation in insulin-dependent diabetes mellitus. Interaction of anti-insulin antibodies and impaired glucagon and epinephrine secretion. Diabetes. 1983 Feb;32(2):134-41.

  6. ^ Segel SA et al. Hypoglycemia-associated autonomic failure in advanced type 2 diabetes. Diabetes. 2002 Mar;51(3):724-33.


Nobody has commented on this article

Commenting is only available for registered Diapedia users. Please log in or register first.