Case Vignette
SI is a 35-year-old woman with past medical history significant for type 1 diabetes mellitus who has been out of insulin for two days and now presents to the emergency department in mild diabetic ketoacidosis (DKA). The Intensive Care Unit (ICU) is at capacity, and the team is discussing treatment options that could be initiated and would not require admission to the ICU.
Explore This Issue
ACEP Now: January 2026When?
Subcutaneous rapid-acting insulin, such as insulin lispro, is a reasonable alternative to intravenous insulin and can be considered in patients with mild-to-moderate DKA, often defined as blood glucose level (BGL) >250 mg/dL, pH >7.0, HCO3 >10 mmol, positive ketones, and an anion gap >12.
Why?
Available literature indicates that insulin therapy is effective for treatment of DKA regardless of the route of administration (intravenous versus intramuscular versus subcutaneous). Traditionally, intravenous insulin has been the preferred treatment method for DKA given concerns about delayed response to subcutaneous insulin. However, with the development of rapid-acting insulin, the subcutaneous route has provided an attractive alternative to treatment of mild-to-moderate DKA, with potential cost savings.
How?
- Dose: The optimal dose in DKA has not yet been established, but initial doses of insulin lispro 0.1-0.3 units/kg subcutaneously have been studied and shown to be effective in this patient population.
- Onset: <15 minutes
- Peak effect: 5-1.5 hours
- Duration: 5 hours
- Adverse Effects: hypoglycemia, hypokalemia, injection site reactions
Overview of Evidence
| Study | Design/Sample Size | Intervention & Comparison | Outcome |
|---|---|---|---|
| Griffey, et al.; 2023 | Prospective pre-post study
N = 177 |
Subcutaneous insulin lispro (n = 78): 0.2 units/kg q2h until BGL <250 mg/dL, then 0.1 units/kg q2h until resolution vs Intravenous insulin infusion (n = 99): Initial bolus of 0.1 units/kg x 1, then 0.1 units/kg/hr titrated per protocol | ED length of stay (LOS) was significantly shorter for the subcutaneous insulin cohort compared to the intravenous insulin infusion group
No differences in hypoglycemic events |
| Karoli, et al.; 2011 | Prospective, randomized trial of patients with mild to moderate DKA
N = 50 |
Subcutaneous insulin lispro (n = 25): 0.3 units/kg x 1, then 0.2 units/kg q2h until BGL <250 mg/dL, then 0.1 units/kg q2h until resolution vs Intravenous insulin infusion (n = 25): Initial bolus of 0.1 units/kg x 1, then 0.1 units/kg/hr until BGL <250 mg/dL, then 0.05 units/kg/hr until resolution | No significant difference in: – Duration of therapy until BGL <250 mg/dL and resolution of DKA – Amount of insulin until BGL <250 mg/dL and resolution of DKA – LOS – Hypoglycemic events – Mortality – Recurrence of ketoacidosis |
| Ersoz, et al.; 2006 | Prospective, randomized trial of patients with mild or moderate DKA defined as BGL >250 mg/dL, beta-hydroxybutyrate level >1.6 mmol/L, pH <7.3, HCO3 level <25 mmol, and positive urine ketones N = 20 | Subcutaneous insulin lispro (n = 10): 0.15 units/kg IV x 1, followed by 0.075 units/kg subcutaneous q1h until resolution
vs Intravenous insulin infusion (n = 10): 0.15 units/kg IV x 1, followed by IV insulin infusion until resolution |
There was no significant difference in: – Time to BGL <200 mg/dL, pH <7.3, HCO3 <18 mEq/L, beta-hydroxybutyrate <0.6 mmol/L, urine ketones negative – Total amount of insulin delivered until resolution of DKA – Hypoglycemic events – Mortality – Recurrence of ketoacidosis |
| Umpierrez, et al.; 2004 | Prospective, randomized trial of patients with DKA (BGL >250 mg/dL, HCO3 <15 mEq/L, pH <7.3, Positive ketones, and beta-hydroxybutyrate level >31 mg/dL) N = 40 |
Subcutaneous insulin lispro (n = 20): 0.3 units/kg x 1 followed by 0.1 units/kg q1h until BGL <250 mg/dL, then 0.05 units/kg q1h until resolution of DKA vs Intravenous insulin infusion (n = 20): Initial bolus of 0.1 units/kg, followed by continuous infusion at 0.1 unit/kg/hr until BGL <250 mg/dL, then 0.05 units/kg/hr until resolution of DKA | No significant difference in: – hospital LOS – duration until resolution of hyperglycemia – duration until resolution of DKA – amount of insulin until resolution of DKA – Episodes of hypoglycemia during therapy – Recurrence of DKA Treatment of DKA in a non-ICU setting was associated with a 39% lower hospitalization charge compared to that in an ICU setting with intravenous insulin infusion |
References
- Micromedex 2.0 (Healthcare Series), (electronic version). Truven Health Analytics, Greenwood Village, Colorado, USA. Available at: http://www.micromedexsolutions.com/.
- Morello C. Pharmacokinetics and Pharmacodynamics of insulin analogs in special populations with type 2 diabetes mellitus.
- Miller K. Subcutaneous Insulin Lispro for Treating DKA.
- Umpierrez G, et al. Efficacy of subcutaneous insulin lispro versus continuous intravenous regular insulin for the treatment of patients with diabetic ketoacidosis.
- Ersoz H, et al. Subcutaneous lispro and intravenous regular insulin treatments are equally effective and safe for the treatment of mild and moderate diabetic ketoacidosis in adult patients.
- Karoli R, et al. Managing diabetic ketoacidosis in non-intensive care unit setting: Role of insulin analogs.
- Griffey R, et al. The SQuID protocol (subcutaneous insulin in diabetic ketoacidosis): impacts on ED operational metrics.
No Responses to “Subcutaneous Insulin for Diabetic Ketoacidosis in Adults”