My daughter recently tried the new CGMS (Continuous Glucose Monitoring System) for three days. We saw that her sugar rises on its own (without eating food or glucose) when she’s been low, and then can stay high for a long time, in spite of correction boluses. Why, if the liver releases sugar in response to a low, do we have to treat lows? Why do some people have seizures and black-outs from lows if the sugar rises without treatment? Should we really fear nighttime lows? I think this is important, because in fear we can easily, accidentally, overtreat lows at night. This, in addition to the rebound hyperglycemia effect results in hours and hours of high blood sugar, which undoubtedly leads to higher HbA1c levels, even though meter averages may not be too bad. Is anyone accumulating data from CGMS usage so that we may know more about exactly how to treat lows to prevent the (sometimes) ensuing highs? Three days is not really enough to establish more than the most obvious trends.

To answer your specific questions:

You do indeed need to treat hypoglycemia. because of the risks of permanent brain damage in infants and young children, and because of the risks of injury from sudden loss of consciousness in young adults, particularly, and for everyone there is the unpleasantness of the symptoms. You do, however, need to think of this primarily in terms of prevention which involves developing a profile of blood sugars for the whole 24 hours and learning how to modify diet and insulin type and dose in response to stress, exercise, and appetite. Carbohydrate supplements in various forms, as well as glucagon, of course, have a role, especially when you know that the blood sugar is low. However, as you point out, it is easy to overtreat, especially if you rely on symptoms only. These principles also apply to nocturnal hypoglycemia.

It has always been supposed that the variation in symptoms as a result of hypoglycemia is due to genetically determined individual variation in the ability of the central nervous system to use other intermediary metabolites such as beta hydroxybutyrate instead of glucose as a source of energy. This has been often shown in laboratory animals, but, for obvious reasons, has not been confirmed in man.

Your short exposure to continuous monitoring must have shown you that we are on the brink of a time when it will be possible, with these devices, to achieve almost normal blood sugar levels. In time, I am sure that electronic sophistication will make it possible to combine glucose sensors and insulin pumps in a way that will safely and effectively mimic the beta cells, but this is some years away. In the meantime, care team members, and people with diabetes and their families, perhaps aided by computer software, are going to have to learn how to use this huge amount of additional information on blood sugar levels to maximal effect. Programs are already available to at least collate information, both at home, or centrally in a doctor’s office, but translating all of this into insulin dose is still in its infancy.

DOB