Utilizing serum bicarbonate instead of venous pH to transition from intravenous to subcutaneous insulin shortens the duration of insulin infusion in pediatric diabetic ketoacidosis.

Background Standard therapy of diabetic ketoacidosis (DKA) in pediatrics involves intravenous (IV) infusion of regular insulin until correction of acidosis, followed by transition to subcutaneous (SC) insulin. It is unclear what laboratory marker best indicates correction of acidosis. We hypothesized that an institutional protocol change to determine correction of acidosis based on serum bicarbonate level instead of venous pH would shorten the duration of insulin infusion and decrease the number of pediatric intensive care unit (PICU) therapies without an increase in adverse events. Methods We conducted a retrospective (pre/post) analysis of records for patients admitted with DKA to the PICU of a large tertiary care children's hospital before and after a transition-criteria protocol change. Outcomes were compared between patients in the pH transition group (transition when venous pH≥7.3) and the bicarbonate transition group (transition when serum bicarbonate ≥15 mmol/L). Results We evaluated 274 patient records (n=142 pH transition group, n=132 bicarbonate transition group). Duration of insulin infusion was shorter in the bicarbonate transition group (18.5 vs. 15.4 h, p=0.008). PICU length of stay was 3.2 h shorter in the bicarbonate transition group (26.0 vs. 22.8 h, p=0.04). There was no difference in the number of adverse events between the groups. Conclusions Transitioning patients from IV to SC insulin based on serum bicarbonate instead of venous pH led to a shorter duration of insulin infusion with a reduction in the number of PICU therapies without an increase in the number of adverse events.