Q: I was wondering whether you could have someone discuss "type 1.5."

I am especially interested in how diabetes "type" and C-peptide numbers are related. I was 28 when I first started having symptoms of diabetes (thirst, weight loss, fatigue). Two years later, I was diagnosed with type 1 diabetes. After a year or two, some confusion arose about what type of diabetes I had. (I needed a rather low total daily insulin dose of 25 units for my weight of 140.) A C-peptide test was run. My blood sugars were kind of high at the time, and the C-peptide came back as 0.2. The doctor told me I was definitely type 1. I am still requiring rather low doses of insulin (a daily total of 30 units for a weight of 170).

One month ago, another C-peptide test was done. My blood glucose was 56 mg/dl at the time, and the test came back at 3.6. It seems to me that my body makes a lot of insulin when I don't really need it and hardly any when I do need it. My blood glucose has been brittle ever since diagnosis.

I wonder what my varying C-peptide results might mean and whether that is the reason I am brittle. Also, what C-peptide numbers does a doctor typically use to determine whether a person is type 1, 2 or "1.5"?

Rebecca
Cypress, Texas

A: This is a very complex topic, primarily because there is no agreement on how to classify people who don't meet the classic description of type 1 and type 2. Now that we have new and better ways to determine diabetes etiology, I suspect it will get better over time.

The child who presents with keto-acidosis and the 50-year-old obese woman who presents with type 2 diabetes after having gestational diabetes during her last pregnancy are easy diabetes types to classify. We are learning that not everyone is "classic", however.

Over the past few years, two terms have entered the literature: "type 1.5" diabetes and latent autoimmune diabetes of adults (LADA). The medical literature on these two types suggests they are one and the same. However, even though there is no consensus on this point, I look at these two groups of patients as very different.

As LADA was initially described in 1994, the adult patients were relatively thin and really looked more like the classic type 1 patients than the obese type 2s. Clinically, we have learned that they are very insulin-sensitive and make insulin for many years—much longer than the typical "honeymoon" period we see in children. In the Diabetes Control and Complications Trial (DCCT), which was mostly young adults, many were still making a bit of insulin at the beginning of the study. C-peptide levels remained higher in those in the intensive therapy group (A1C about 7%) compared to standard therapy (A1C about 9%). This is the best evidence that tight control preserves insulin secretion in these adults with newly diagnosed type 1 diabetes.

It is also critical to appreciate that LADA is just a subtype of type 1 diabetes, meaning that LADA patients also have a high frequency of the same autoimmune markers we see in children. These include islet cell antibody (ICA), glutamic acid decarboxylase (GAD), islet cell antibody-512, and insulin autoantibody. So patients with LADA may make C-peptide for a very long time and have at least one of these antibodies in their blood. Unfortunately, over time these antibodies often tend to go away.

My favorite story concerns a lady in her 50s on only 15 units of insulin. Her doctor started her on insulin the day she was diagnosed with diabetes. Her GAD antibody was positive when I saw her, but her diabetes had been diagnosed 31 years previously! So measuring C-peptide may not be the best way to understand the type of diabetes. Really, the entire picture needs to be looked at.

I see "type 1.5" diabetes as being very different. These are people who look like type 2 patients (obese, often with high blood pressure, high triglycerides, low HDL cholesterol), yet when the antibodies mentioned above are measured, one comes back as positive, suggesting that there is also an autoimmune process going on leading to insulin deficiency. We don't yet know the best way to treat these folks, but they appear to require insulin much sooner than the type 2s without the antibody. Still, for a while they do respond to insulin sensitizers such as Glucophage (metformin), Actos (pio-glitazone) and Avandia (rosiglitazone).

This confusing picture can be summarized as follows: LADAs look like type 1s (generally thin) and are insulin-sensitive, whereas "type 1.5s" look like type 2s and are insulin-resistant. Both have one or more antibodies that are positive. This is, of course, my interpretation of what I see in my clinic; but to be fair, there isn't enough published research for everyone to agree with what I see.

To take it one step further, we also see classic type 1s from childhood who later develop insulin resistance, usually from an inherited defect (family history of type 2 diabetes) but occasionally from polycystic ovary syndrome (PCOS) in women who are insulin-resistant. So here, the type 1 starts first, and then they develop obesity and insulin resistance. I say that those folks have "type 3" diabetes, although others refer to them as people with "double double" diabetes or "hybrid" diabetes.

As for the question about C-peptide tests, it is good to know the glucose level when measuring the C-peptide. With a low glucose reading, however, we would expect a low C-peptide, so the high C-peptide seen with the glucose reading of 56 makes me wonder how good the assay for C-peptide was. We've learned over the years that not all labs do a good job in measuring this.

Irl B. Hirsch, MD
Professor of Medicine, University of Washington School of Medicine
Seattle, Washington