Our 16 1/2 year old daughter was diagnosed with type 1 diabetes almost four years ago. At that time, she had a blood sugar of 434 mg/dl [24.1 mmol/L], but no ketones, and over the years her A1cs have all been normal with only a few scattered high blood sugars. For the first few years, she was on Humalog and NPH, but was having lots of lows so we switched to Lantus about a year ago with no shots before meals which helped her lows.

At our last visit the new care giver was suspicious that she may not have diabetes, ordered antibody tests, and the “main two” were negative. We were told to stop her insulin and that she may not have diabetes. For the last four days her blood sugars have been within normal limits without insulin so obviously we are concerned.

I understand your concerns. Without complete access to all the information, data, and health records, I can only offer some general information in return.

As you probably know or have seen on this web-site, not all elevated blood glucoses are due to diabetes, not all children with diabetes have type 1, and not all people with type 1 have detectable antibodies. There are certain ethnic groups with obvious “insulin-dependency” who tend to have antibodies less often, including African-Americans and Latinos. After four years of carrying a diagnosis of diabetes and being treated with insulin, I do not know that I would put too much stock in the presence of insulin antibodies. I presume the “main two” that you refer are the GAD 65 and ICA 512 antibodies. If her glucose readings have “always been normal” (and I mean always less than 125 mg/dl [mmol/L] before meals), and her hemoglobin A1c has always been normal (depending on the lab, the upper limits can vary – so you cannot ideally compare A1c values from different labs) then I could well support a cautious, careful trial off of insulin.

Is your daughter overweight? Could she have had type 2 diabetes, but because of her age, was labeled as a probable Type 1? Maybe. Not all people with type 2 are heavy, but most are, even in a young person. However, type 2 diabetes in a then 12 year old is certainly less common than type 1.

If she has never had ketones associated with elevated blood glucose, then I think a trial off insulin is reasonable if done carefully. Personally, I would strongly consider doing a C-peptide level and would even consider performing a glucose tolerance test.

An oral glucose tolerance test must be done most correctly (dose of 1.75 grams of glucose to a maximum of 75 grams; high carb loads to comprise about 60% of calories on the three days before the test; measure INSULIN as well as glucose). Indeed, this might be a young person in whom a very specially arranged intravenous glucose tolerance test might be helpful.

One could argue that she has simply experienced a prolonged honeymoon phase of Type 1 diabetes. The longest I have personally encountered was just shy of four years. However, even “honeymooners” do not have glucose readings that are always normal so I think that point needs verification.

DS