MKSAP quiz: Ketoacidosis at diabetes diagnosis

This month's quiz asks readers to evaluate a 50-year-old man who was hospitalized for diabetic ketoacidosis 3 months ago but had no previous history of diabetes mellitus.


A 50-year-old man is evaluated during a follow-up visit for diabetes mellitus. He was hospitalized for diabetic ketoacidosis 3 months ago. He had no previous history of diabetes mellitus. His blood glucose levels since discharge have ranged from 90 to 120 mg/dL (5.0-6.7 mmol/L). Medications are glargine insulin and lispro insulin with meals.

On physical examination, vital signs and all other findings are normal.

Which of the following is the most appropriate management?

A. Continue insulin and add metformin
B. Discontinue insulin and monitor
C. Discontinue insulin and start metformin
D. Measure fasting C-peptide and glucose levels

Reveal the Answer

MKSAP Answer and Critique

The correct answer is D. Measure fasting C-peptide and glucose levels. This item is available to MKSAP 18 subscribers as item 2 of extension set 4 in the Endocrinology and Metabolism section. More information about MKSAP 18 is available online.

The most appropriate management of this patient's ketosis-prone diabetes (KPD) is to measure fasting C-peptide and glucose levels. KPD encompasses a heterogeneous syndrome characterized by the onset of diabetic ketoacidosis (DKA) in persons who do not have the typical phenotype of type 1 diabetes. These individuals may experience relative or absolute insulin deficiency during the time of DKA followed by periods of continued insulin dependence or independence. KPD can be characterized by the patient's autoantibody status (for example, glutamic acid decarboxylase [GAD65] and tyrosine phosphatases IA-2 antibodies) and the pancreatic β-cell function. This information is helpful in determining the patient's long-term clinical course and need for insulin. Fasting C-peptide or glucagon-stimulated C-peptide testing can be used to assess β-cell function. Individuals without β-cell function (as indicated by low or absent C-peptide level), regardless of the autoantibody status, will require lifelong insulin therapy. The patient had little to no β-cell function at the time of DKA, which necessitated treatment with exogenous insulin. As the glucotoxicity from DKA resolves with improved glycemic control, there is the possibility of partial β-cell recovery. Given that the patient's current blood glucose values are now controlled, C-peptide should be measured to determine his β-cell function to help guide future therapeutic decisions regarding insulin use.

The patient's blood glucose values are well controlled on his current regimen, so the addition of metformin is unnecessary. However, if further testing reveals that the patient has evidence of remaining β-cell function (particularly in the absence of antibodies), then an attempt could be made to taper insulin and initiate metformin.

Before discontinuing insulin or substituting metformin for insulin, evaluation is required to determine whether β-cell function remains, as this will help guide future decisions regarding the need for insulin as part of the regimen. If the patient has no evidence of β-cell function, discontinuing insulin increases his risk for developing DKA again.

Key Point

  • Fasting C-peptide or glucagon-stimulated C-peptide testing can be used to assess β-cell function, which is helpful in determining the long-term clinical course and need for insulin in patients with ketosis-prone diabetes.