• acidosis;
  • canine;
  • diabetes mellitus;
  • ketosis



To characterize the use of lispro insulin in dogs with diabetes ketoacidosis (DKA) and to compare the length of time required for resolution of hyperglycemia, ketosis, and acidosis, respectively, in dogs with DKA treated with lispro or with regular insulin.


Randomized prospective clinical trial performed between November 2006 and May 2009.


University teaching hospital.


Client-owned dogs with naturally occurring DKA. Dogs with a blood glucose (BG) > 13.9 mmol/L (>250 mg/dL), blood pH between 7.0 and 7.35, and a blood beta-hydroxybutyrate (BOHB) concentration >2.0 mmol/L were eligible to be enrolled into the study and were randomly assigned to receive an IV continuous rate infusion (CRI) of either lispro or regular insulin.


Lispro or regular insulin was administered as an IV CRI at an initial dose of 0.09 U/kg/h. The dose was adjusted according to a previously published protocol.

Measurements and Main Results

Twelve dogs were enrolled into the study. The time to biochemical resolution of DKA was defined as the time interval from when the IV CRI of insulin began until marked hyperglycemia (BG > 13.9 mmol/L [>250 mg/dL]), acidosis (venous pH < 7.35), and ketosis (BOHB concentration >2.0 mmol/L) resolved. The median time to biochemical resolution of DKA in dogs treated with lispro insulin was significantly shorter (26 h; range 26–50 h) than in dogs treated with regular insulin (61 h; range, 38–80 h, P = 0.02). Median admission blood glucose concentration of all 12 dogs (24 mmol/L [432 mg/dL; range, 17.8–38.9 mmol/L [321–700 mg/dL]) decreased significantly with fluid resuscitation and prior to insulin therapy (20.5 mmol/L [369 mg/dL; range, 14.5–33.3 mmol/L [261–600 mg/dL], P = 0.0085). No adverse effects were observed in association with IV lispro insulin administration.


Treatment of DKA in dogs with IV CRI lispro insulin is safe, and as effective as treatment with regular insulin.