The goal of this study is to quantify day-to-day changes in blood glucose during treatment towards remission in ketosis-prone diabetes (KPDM) and describe them using a mathematical model of KPDM pathogenesis and remission.

Approximately half of newly diagnosed obese African Americans presenting with diabetic ketoacidosis (DKA) have clinical, metabolic and immunologic features of type 2 diabetes (T2D), also known as ketosis-prone diabetes (KPDM). Unlike patients with type 1 diabetes, with intensive insulin treatment, approximately 70% of patients with KPDM exhibit improved pancreatic β-cell function and insulin sensitivity to allow discontinuation of insulin therapy (indicating near-normoglycemia remission based on fasting blood glucose (FBG) < 130 mg/dl and HbA1c < 7% off insulin therapy during follow-up]). The clinical course of KPDM is variable, with the duration of remission ranging from 6 to 120 months. The origins of this variation in the ability to achieve remission and its duration are poorly understood.

It has been observed that a 20-hour infusion of glucose reduces pancreatic beta (β)-cell function in a KPDM patient, and ketotic relapse is often preceded by hyperglycemia. The researchers thus hypothesized that the differences between conventional T2D and KPDM may be explained by the presence of a reversible glucotoxicity process, which operates on the timescale of days. The researchers developed a preliminary mathematical model describing the pathogenesis and remission of KPDM using such a process. The researchers showed that, by changing the rate of this hypothesized glucotoxicity process, this model can produce a variety of clinical courses, describing both conventional T2D and KPDM with varying rates and durations of remission.

This study is a pilot study that will refine and validate this model using prospective clinical data with continuous glucose monitor (CGM) data from patients with KPDM.

Patients will receive standard-of-care treatment for their diabetes as per their treating physician. Insulin therapy is the standard of care after an episode of DKA. Therefore, all participants will be discharged on insulin. A CGM will be placed on these participants at discharge from the hospital until insulin discontinuation. Since many of these patients insulin needs decrease after discharge from the hospital, the study team will utilize CGM glucose readings to adjust insulin doses.

Inclusion Criteria:

• Provide informed consent
• Have a BMI ≥ 28 kg/m2
• Be of African American ancestry
• Meet diagnostic criteria for DKA. Diagnostic criteria for DKA will include a plasma glucose > 250 mg/dl, a venous pH < 7.30, a serum bicarbonate < 18 mmol/l, and serum ketones (beta-hydroxy butyrate) > 1.5 mmol/L.

Exclusion Criteria:

• Significant medical or surgical illness including but not limited to myocardial ischemia, congestive heart failure, chronic peripheral venous insufficiency, chronic renal insufficiency, liver insufficiency (serum transaminases 3 times the upper limit of normal) and acute or chronic infectious processes
• Have recognized uncontrolled endocrine disorders such as hypercortisolism, acromegaly, or hyperthyroidism
• Anemia (hemoglobin < 12.5 g/dL for men, <11.5 gm/dL for women), bleeding disorders, or abnormalities in coagulation studies
• Pregnant
• Diagnosis of diabetes > 90 days before the presentation of DKA
• Unable to give consent