Should I rescue a patient with an automated insulin pump (AIP) and potential diabetic ketoacidosis (DKA) by suspending the pump and initiating intravenous insulin infusion?

Management of Automated Insulin Pump in Potential DKA

If you suspect DKA in a patient with an automated insulin pump, immediately discontinue the pump and start intravenous insulin infusion—do not rely on the pump to manage a hyperglycemic crisis. 1

Immediate Actions for Suspected DKA

Disconnect the Pump Immediately

• Disconnect the pump from the insertion cannula but leave the pump running to avoid inadvertently stopping it completely if you're unfamiliar with the device interface 1
• Start IV insulin infusion immediately upon disconnection—patients become relatively insulin deficient within one hour of pump removal 1
• Ideally, begin IV insulin at least 30 minutes before removing the pump to prevent gaps in insulin coverage, though in emergency situations this may not be possible 1

Contraindications to Continued Pump Use

The pump must be discontinued in the following situations:

• Hyperglycemic crises (DKA or HHS) 1
• Impaired level of consciousness 1
• Patient inability to correctly use pump settings or self-manage diabetes 1
• Lack of trained healthcare providers or hospital policies supporting pump use 1

IV Insulin Protocol for DKA

Initiation

• Continuous intravenous insulin infusion is the preferred method for DKA management unless the episode is mild 1
• Prime IV tubing with 20 mL of insulin solution before starting the infusion to minimize insulin adsorption to plastic tubing 2
• Use the patient's hourly basal rate from their pump settings as an initial guide for IV insulin rate, then titrate to blood glucose 1

Monitoring

• Check blood glucose every 2-4 hours during DKA treatment 1
• Monitor serum electrolytes, venous pH, and anion gap every 2-4 hours 1
• Venous pH (usually 0.03 units lower than arterial pH) is adequate for monitoring—repeat arterial blood gases are generally unnecessary 1

Resolution Criteria

DKA is resolved when:

• Glucose <200 mg/dL 1
• Serum bicarbonate ≥18 mEq/L 1
• Venous pH >7.3 1

Transitioning Back to Pump Therapy

When Patient is Stable

• Reconnect the pump and allow it to infuse at the basal rate for at least 2 hours before stopping IV insulin to establish adequate subcutaneous insulin depot 1
• Continue hourly glucose monitoring during the transition period 1
• This overlap prevents rebound hyperglycemia and ketogenesis 3

If Pump Cannot Be Restarted

• Calculate the 24-hour total basal dose from pump settings 1, 4
• Replace with long-acting insulin (glargine or detemir) in 2 divided doses 12 hours apart 1, 4
• Discontinue pump 2 hours after first injection of basal insulin 1, 4
• Maintain same insulin-to-carbohydrate ratios for mealtime boluses 1, 4

Critical Pitfalls to Avoid

Rapid Metabolic Decompensation

• Never delay IV insulin while attempting to troubleshoot the pump—disconnection from a pump renders patients relatively insulin deficient within one hour 1
• The risk of rapid hyperglycemia and ketosis upon pump cessation is significant 1

Inadequate Transition

• Abrupt discontinuation of IV insulin without adequate subcutaneous insulin overlap leads to poor glycemic control 1
• Ensure 1-2 hour overlap when transitioning from IV to subcutaneous insulin 1

Monitoring Errors

• Do not use nitroprusside method for ketone monitoring during treatment—it doesn't measure β-hydroxybutyrate (the predominant ketone in DKA) and may falsely suggest worsening ketosis 1
• Blood β-hydroxybutyrate measurement is the preferred monitoring method 1

Special Considerations for Automated Insulin Delivery Systems

• Newer hybrid closed-loop systems (Medtronic 670G, Tandem Control-IQ, Diabeloop) have shown promise in non-critical inpatient settings, but should not be used during hyperglycemic crises 1
• Automatic threshold suspend features should be turned off in the hospital setting 1
• Research on automated insulin delivery in hospitalized patients is limited to stable, non-critical situations 1