About five months after diagnosis of type 1 diabetes a year ago, my three year old son still had highs and lows, but we assumed that he was out of his honeymoon period, and he was under pretty good control. At seven months after diagnosis, he was having so many lows (especially at night) that we started to decrease his insulin. There are times at bedtime or in the middle of the night that he is as high as 450 mg/dl [25 mmol/L], but by morning his blood sugar is about 160 mg/dl [8.9 mmol/L]. At first we thought he was still honeymooning, but it has been a year and there is no sign of this changing. He's having a lot of lows again even though he is on only one injection of NPH with Humalog in the morning. I'll discuss decreasing his insulin some more when we see his endocrinologist in a few weeks. If my son is making insulin, is it at all possible that he could benefit from an oral hypoglycemic agent? How do I know for sure he has type 1 diabetes? Is there a way to determine if he has any islet cells left?

The diagnosis of type 1A (autoimmune) diabetes (which is probably what your small son has) is based on a positive antibody test which is now usually done at the time of diagnosis. In a small percentage of Caucasian children and in about half the new onset cases in African American and Hispanic children, this test is negative, and the diagnosis is then usually type 1B diabetes. The distinction is of some importance because, in the 1B form, it is often possible to gradually dispense with insulin after a few weeks or months, although the expectation is that insulin dependance will return in time. Glucophage [metformin] has been used successfully to improve control in some cases.

in the case of your son, I think his insulin regimen needs to be reviewed with some urgency. Random blood sugars of 450 mg/dl [25 mmol/L] and fasting morning blood sugars of 160 mg/dl [8.9 mmol/L] are not normal. When mixed with abnormally low levels they may yield a hemoglobin A1c level that offers the illusion of satisfactory control.

My suggestion is that you talk to the endocrinologist about using Lantus (insulin glargine) for 24 hour basal insulin needs and a new fast -acting insulin (Humalog or NovoLog) given after meals so as to modify the dose according to the amount of ‘carbs’ actually consumed as well as the premeal blood sugar. The use of glargine certainly reduces the incidence of nighttime lows. This of course means four injections instead of one, and, even if you use ultrafine needles this may be hard on a small child. The glargine has to be given separately but can be given in the morning at the same time as the fast -acting insulin which can be mixed with NPH so that a separate injection is not needed at lunch time. You might even discuss the possibility of using an insulin pump even at this age, although I think it would be much better to try a changed insulin regimen first.

DOB