Atypical diabetes: Case study 3
A 26-year-old African–American male presents to the emergency room (ER) with 24 hours of nausea and vomiting. He reports that, while he is normally healthy, he has been losing weight and has not been well for about a month. The patient was safe to lose some weight (as he needed to) but he thought that he was losing more than he should have been. Over this past weekend, he started to have an upset stomach and then, yesterday, he started to feel nauseated. Finally, he started vomiting last night and thought that he had gotten food poisoning or maybe a stomach virus. The vomiting got worse over the course of today so his family brought him to the ER. He has not been in contact with anybody who was ill, has not recently traveled, and is not sexually active. Box 1 details the findings of his initial medical examination.
|Findings from the patient’s initial medical examination|
The patient is laying on the gurney and appears to have stomach pain, is tachypneic with Kussmaul breathing, and has dry oral mucosa. He has some abdominal discomfort to light palpation, but no focal pain, and does not appear to have an acute abdomen. Other than his obesity, the rest of his exam is normal. You obtain his initial laboratory test results, detailed in Box 2.
|Findings from the patient’s laboratory tests|
Question 1. Based upon his presentation to the ER, which of the following diagnoses are you likely to make for his admission?
- New-onset diabetes mellitus
- Acute diabetic ketoacidosis
- Dehydration (prerenal azotemia)
- All of the above
His family comes to the ER shortly after and want to know more.
Question 2. What type of diabetes does the patient likely have?
- Type 1 diabetes
- Type 2 diabetes
- Latent autoimmune diabetes of the adult (LADA)
- Monogenic diabetes/maturity onset diabetes of youth (MODY)
- Ketosis-prone type 2 diabetes
- Cannot be sure at this time
Question 3. What laboratory tests would you order at this time to confirm the patient’s diabetes type? Choose all options that apply.
- Oral glucose tolerance test
- C-peptide and glucose test
- Glutamic acid decarboxylase (GAD) and islet cell autoantibodies
- Lipid panel
The patient is in diabetic ketoacidosis (DKA) and this is the first and most important thing to treat and stabilize. It may actually be difficult to determine what type of diabetes he has at this time since most of the tests we would want to order will not be accurate or will take too long to get back to be useful for the acute treatment.
He was treated using the DKA protocol (ie, fluids, intravenous insulin, and electrolyte supplementation) and his condition improved over the next 36 hours. It was clear that his insulin needs were significant. Once his DKA was resolved, he was started on subcutaneous insulin, requiring 48 units of basal insulin glargine and 14 units of insulin aspart at each meal, and a correction of 3 per 50 units above 150 units.
When we spoke to his family on the first evening, we told them that he definitely had diabetes, but that we were not yet clear what type. The fact that he was young and in DKA made us think he had type 1 diabetes. However, his obesity, family history, and the need for a lot of insulin during treatment made us consider type 2 diabetes. Unfortunately, the normal testing we would have done to measure his endogenous insulin production (C-peptide coupled with a glucose test) would have been inaccurate during his initial presentation as his severe hyperglycemia would have suppressed endogenous glucose production (glucotoxicity). Therefore, we told the family that we would have to treat him with insulin, as if he had type 1 diabetes, and that once he was stable we could order additional testing to make an accurate diagnosis.
He was in the hospital for 5 days and was sent home on insulin, which he did quite well with. His glucose normalized within a couple of weeks. He did have additional testing done and, in the stable outpatient setting, his C-peptide was actually high, with a blood glucose of 160 mg/dL, and his autoimmune testing did not show any GAD antibodies or islet cell autoantibodies. His lipids were also ordered and showed classic diabetic dyslipidemia, with a total cholesterol of 248 mg/dL, high-density lipoprotein of 32 mg/dL, low-density lipoprotein of 140 mg/dL, and triglycerides of 220 mg/dL.
A review of ketosis-prone type 2 diabetes
There is a subpopulation of people with type 2 diabetes who will develop DKA. This may only be 5–10% of all people with type 2 diabetes, but their presence is important [1, 2]. Ketosis-prone type 2 diabetes is much more common in Africa and Latin American countries, and is much more common in men . DKA is the most common complaint presenting for the diagnosis of diabetes . Most people with ketosis-prone type 2 diabetes can stop insulin within 3 months of DKA and many can be managed with oral medications or lifestyle changes alone .
One way to determine the diabetes type of a person has who has been in DKA is to use the AB system (see Box 3). This tests the person once they are stable following a DKA episode and measures both antibodies (A) and endogenous beta-cell function (B) . You may also be able to clinically determine a patient’s diabetes type from their family history, associated metabolic conditions (eg, dyslipidemia), and sensitivity to insulin.
|Box 3. The AB classification system |
|A-B+||These patients are mostly like to have ketosis-prone type 2 diabetes. They have preserved beta-cell function and lack markers of islet cell autoimmunity. Most can discontinue exogenous insulin therapy.|
|A+B-||These patients have type 1 diabetes, requiring lifelong exogenous insulin therapy.|
|A-B-||These patients probably also have type 1 diabetes. We would classify them as type 1B. They have permanent and complete beta cell failure, but lack serologic markers of islet cell autoimmunity. It was recently discovered that zinc transporter antibodies are present in these patients. They require lifelong exogenous insulin therapy. |
|A+B+||This is a challenging group. They may have latent autoimmune diabetes of the adult (LADA), but it would be unusual to have preserved beta-cell function once they went to diabetic ketoacidosis. LADA would be supported if they were phenotypical of type 1 diabetes, they did not have diabetic dyslipidemia, and they were sensitive to insulin. We do know that some people can have markers of autoimmunity but actually have no diagnosis of diabetes.|
Remember that there are many types of diabetes. The patient described here looked like he had type 2 diabetes, both phenotypically and by their family history. However, he presented in DKA at a young age and we had to stabilize him through acute treatment to be able to determine his insulin sensitivity. Once stable, it was clear that he had type 2 diabetes. It is important to remember that people with type 2 diabetes can develop DKA, and, in fact, this subgroup is prone to DKA. Prevention, self-care, and prompt intervention is critical when a patient is sick.
Take home points
- When in doubt, always treat the patient as if they have type 1 diabetes.
- People with type 2 diabetes can go into DKA.
- Do not test endogenous insulin production when the patient is acutely glucotoxic.
- People with ketosis-prone type 2 diabetes should be counseled on their condition to prevent DKA and to inform treating facilities, since it will affect their treatment.
- Guillermo E. Umpierrez, Dawn Smiley, Abbas E. Kitabchi; Narrative review: Ketosis-prone type 2 diabetes mellitus. Ann Intern Med 2006;144:350–357.
- Umpierrez, GE. Ketosis-prone type 2 diabetes: Time to revise the classification of diabetes. Diabetes Care 2006;12:2755–2757.
- Smiley D, Chandra P, Umpierrez GE. Update on diagnosis, pathogenesis and management of ketosis-prone Type 2 diabetes mellitus. Diabetes Manag (Lond) 2011;1:589–600.
- Banerji MA, Chaiken RL, Lebovitz HE. Prolongation of near-normoglycemic remission in black NIDDM subjects with chronic low-dose sulfonylurea treatment. Diabetes 1995;44:466–470.
- Balasubramanyam A, Garza G, Rodriguez L et al. Accuracy and predictive value of classification schemes for ketosis-prone diabetes. Diabetes Care 2006;29:2575–2759.