‘Type 3c diabetes’ may be more common than type 1

Diabetes most often results from one of two conditions. The most common is type 2 diabetes — formerly “adult onset” diabetes, due to a combination of insulin resistance and pancreatic beta cell failure. This accounts for some 26 million people in the United States. The second most common is type 1 diabetes, which accounts for some 2.5 million people in the United States. The older terms “juvenile” — previously used for type 1 diabetes — and “adult onset” — used for type 2 diabetes — have been abandoned because it is clear that adults may develop autoimmune (type 1) diabetes and children may develop insulin-resistant (type 2) diabetes.

The article makes many important points. First, it reminds patients and physicians that there are still other less common forms of diabetes, that despite their relative rarity, are still encountered because of the marked prevalence of all forms of diabetes, a prevalence that is increasing throughout the world. One example is maturity onset of diabetes in the young (MODY), a cluster of diabetes subtypes that share specific genetic defects that lead to inadequate beta cell mass and/or function. Another common example is glucocorticoid drug treatment (ie, prednisone, cortisone), which impairs both insulin action and beta cell survival or function. There are many other rare forms of diabetes. In this report, in an unselected series of 30,000 people recorded in an English public health system database during a 15-month period, the authors identified 559 people whose diabetes could be attributed to preceding pancreatitis or pancreatic surgery, conditions that are associated with beta cell loss and, therefore, inadequate insulin secretion.

One surprising note is that diabetes following pancreatic disease, called “type 3c diabetes” by some, usually went undiagnosed for years, because the onset was in adulthood and was simply assumed to be type 2 diabetes. Distinguishing between type 2 diabetes and pancreatic disease-associated diabetes makes a difference, because the former may be effectively treated with drugs other than insulin, whereas inadequate pancreatic insulin secretion must be managed by insulin treatment. Indeed, the patients with pancreatic diabetes in this series required insulin more often and were evidently undertreated because their HbA1c levels, a marker of good glucose control, were higher than the classical pattern in patients with type 2 diabetes.

Another surprise is that pancreatic disease-associated diabetes in this adult group (2.6 new cases per 100,000 person years) was actually more common than type 1 diabetes (1.6 new cases per 1000,000 person years). Most practicing physicians would probably not have guessed that this would be the case.

Another important parenthetic comment is that this study illustrates the power of single-provider health care systems that permit the rapid collecting of these types of population-based data. This is not currently possible in the United States, except in adults over 65 years covered by Medicare and except in veterans cared for by the VA Health System. These obviously miss a large part of the population.

Finally, readers may be interested to know that in large databases, such as the current report, or large health system databases, such as Medicare and the VA health system, it is not possible to define who precisely has type 1 diabetes and who has type 2 diabetes, because the tests required to define type 1 diabetes (ie, beta cell antibodies, C-peptide measurement and HLA genotypes) are not commonly performed, or may not be in the medical record. Thus, in most health systems as in this study, people are bundled into the type 2 diabetes rubric simply because they are adults, or perhaps adults with high BMI. Conversely, the type 1 diabetes label is attached if they have been assigned a diagnosis of type 1 diabetes as a billing code by a physician, but only a handful have actually been demonstrated by the strict criteria enumerated above to have type 1 diabetes. Thus, one final take-home message from the paper is that in addition to overlooking or mislabeling less common diabetes subtypes such as pancreatitis-related diabetes, the two common subtypes of diabetes are often confused as well. It is always interesting, and often helpful to the patient, to take a deeper diagnostic dive.