• Mean Daily Blood Glucose Concentration While on the Insulin Drip [ Time Frame: blood glucose (BG) before meals and at bedtime ] [ Designated as safety issue: No ]
• Frequency of Hypoglycemia [ Time Frame: blood glucose (BG) before meals, at bedtime and as needed ] [ Designated as safety issue: Yes ]
• Frequency of Hyperglycemia [ Time Frame: blood glucose (BG) before meals, at bedtime and as needed ] [ Designated as safety issue: Yes ]

Diabetic ketoacidosis (DKA) is the most serious emergency in patients with diabetes. With an estimated 100,000 admissions per year in the United States, DKA is also the leading cause of death in children with type 1 diabetes, and accounts for a significant proportion of admissions in adult patients with type 1 and type 2 diabetes. The mainstay in the treatment of DKA involves the continuous intravenous (IV) infusion of regular insulin or the frequent subcutaneous (SC) injections of regular or rapid-acting insulin analogs. Multiple studies have reported successful protocols for insulin administration during the acute management of DKA, but they have failed to address the transition phase from IV to SC maintenance insulin regimen. The American Diabetes Association (ADA) position statement recommends the use of split-mixed insulin combination of regular and intermediate-acting insulin (NPH). This regimen, however, are associated with a high rate of hyperglycemia shortly after discontinuation of IV insulin and a risk of hypoglycemia during the hospital stay. Recently, the long-acting "basal" insulin glargine (Lantus®, Sanofi Aventis Pharmaceuticals) has been shown to facilitate glycemic control with lower rate of hypoglycemic events than intermediate-acting insulin in subjects with type 1 and type 2 diabetes. This study aims i) to determine the effects of giving a dose of glargine insulin shortly after starting an intravenous insulin infusion on glycemic control, time to resolve DKA, and rate of hypoglycemia in patients with DKA, and ii) to compare the safety and efficacy of basal/bolus (glargine/glulisine) insulin versus the standard split-mixed insulin regimen of NPH and regular insulin after the resolution of DKA. The hypothesis is that basal (lantus®) insulin as compared to NPH insulin shortly after the start of insulin infusion will improve inpatient glycemic control in patients with DKA.

This investigator initiated research will be conducted at Grady Memorial Hospital, Atlanta and at University of Minnesota, MN. Dr Umpierrez designed the study and will serve as principal investigator. A total of 40 patients will be recruited at each site.

• Drug: insulin glargine+ glulisine
• Drug: NPH + Regular insulin

• Experimental: Daily insulin glargine (Lantus) + glulisine (Apidra) before meals
• Active Comparator: Split-mixed NPH + Regular insulin twice daily

Inclusion Criteria:

• All patients admitted to Grady Memorial Hospital who meet diagnosis criteria of DKA and who are willing to participate in the study protocol will be considered candidates for inclusion into the study.
• Diagnostic Criteria for DKA: Blood glucose > 250 mg/dL, arterial or venous pH < 7.3, serum bicarbonate < 18 mEq/L, and moderate to severe ketonemia (acetoacetate ≥ 1:4 or β-hydroxybutyrate > 3 mmol).

Exclusion Criteria:

• Hemodynamic instability (MAP < 50 or patients requiring pressor)
• Significant identifiable medical or surgical illness, including but not limited to: acute myocardial infarction, congestive heart failure; respiratory failure requiring mechanical ventilation; acute or chronic renal insufficiency (serum creatinine > 3.0 mg/dl); end stage liver failure, and cirrhosis.
• Patients with dementia or persistent altered mental status that would prevent collection of consent form and reliable information.
• Pregnancy