New Insights into the Management of Diabetic Ketoacidosis
World Small Animal Veterinary Association World Congress Proceedings, 2009
Ricardo Duarte, DVM, PhD
Hospital Veterinário Pompéia, São Paulo, SP, Brazil

The diagnosis of diabetic ketoacidosis (DKA) in the ill-appearing diabetic dog is usually straightforward. Dogs with DKA usually present with dramatic clinical manifestations, such as anorexia, vomiting, and lethargy. Within such a setting, detection of ketonuria can rapidly confirm the suspicion of DKA, but it has few implications in the subsequent approach. One can estimate ketonemia by applying a drop of serum or plasma on the appropriate reagent of the urine test strip, but this test only detects acetoacetate and is a semiquantitative test.

The measurement of plasma beta-hydroxybutyrate (β-OHB) on admission can provide additional information. Based on a previous study, dogs with plasma β-OHB >2.0 mmol/L should receive ambulatory monitoring and treatment, until the results of additional tests. If plasma β-OHB is >3.8 mmol/L the diagnosis of DKA is confirmed and intensive care is warranted.1 Coincidentally, a recent study in human beings with DKA, suggested that the same cutoff value of plasma β-OHB should be used for the diagnosis of DKA, using a portable meter (MediSense Optium, Abbott Laboratories).2 This device has been validated for the use in dogs.3

Evaluation of blood gases and pH is still needed. In a study of our group (unpublished data), mixed acid-base disorders were common, chiefly high anion gap acidosis and concurrent respiratory alkalosis, and hyperchloremic acidosis with moderated to marked increases in serum B-OHB. The implications of these findings are unknown, but in human patients with DKA, both conditions can possibly slower the recovery from metabolic acidosis. Because respiratory alkalosis is the expected physiologic response to metabolic acidosis, this mixed acid-base disorder may be difficult to recognize clinically, and the diagnosis can be made only by assessment of arterial blood gases. Other tests should be part of the initial data base, such BUN, electrolytes and inorganic phosphate, and others can be ordered, based on clinical evaluation.

The treatment of DKA in dogs and cats is based on a few original publications. At our Hospital, insulin infusion at a rate of 0.05 U/kg/h appears to be as effective as the commonly recommended dose (0.1 U/kg/h), with fewer complications, such as hypoglycemia and hypokalemia. Especially in cats, a slow-motion recovery of the metabolic abnormalities seems advisable. Cats are prone to iatrogenic fluid overload and severe hypokalemia. Some cats are also very sensitive to exogenous insulin. In human medicine, there was a strong belief that "DKA is rare in true type 2 diabetes mellitus". This dogma has changed in recent years. And cats with DKA at diagnosis of diabetes also can go into remission, after treatment with insulin.4

Although hypophosphatemia is common and develops during treatment of DKA5, we do not routinely give phosphate treatment because hypocalcemia is also common in dogs with DKA (and phosphate therapy worsens hypocalcemia). Phosphate therapy is advised in patients with initial low inorganic phosphate or those with conditions like anemia, congestive heart failure or other reasons that could be further compromised by hypophosphatemia. Ionized calcium should be measured during phosphate therapy. Also, we do not advocate the use of adjunctive bicarbonate at any pH level.

Other adjunctive medical therapies may be indicated according to the clinical picture. These include antibiotics and analgesics, for example. Vomiting is a common sign associated with DKA. We have success in controlling vomiting with ondansetron (0.5 mg/kg BID or TID).

No matter what protocol one uses, constant monitoring is the key point for the success of the therapy. Monitoring the patient with severe DKA requires close observation because of the evolving nature of the disease and also because treatment can cause other complications.

Blood glucose monitoring is important and the goal is a 50-75 mg/dL reduction per hour. If the reduction is >100 mg/dL/h, too much insulin was given (bad for potassium). If Blood glucose is not declining as expected, urine output and blood pressure must be checked, and fluid therapy must be reevaluated, before doubling the infusion rate of insulin.

Ideally laboratory monitoring should include blood gases and plasma electrolytes, including ionized calcium, but costs can be a limiting factor. Monitoring plasma potassium at least each 4h is important, principally in those patients with low potassium on admission. Frequent clinical assessment is imperative. Some patients will require constant monitoring of blood pressure and urinary output.

Early studies in human beings suggested that de sequential measurement of blood β-OHB, could be used for monitoring treatment. However, Durocher and colleagues demonstrated that β-OHB may not be the predominant ketone body in DKA.6 This finding is the opposite of what occurs in human patients with DKA, who have high β-OHB to acetoacetate ratios. This novel evidence mandates a reappraisal of the use of β-OHB concentrations to assess the degree of hyperketonemia in dogs, but it doesn't invalidate its use as a screening test.

Once the patient is eating spontaneously, the intensive care can be slowly discontinued. At this point, dogs generally have a plasma bicarbonate ~13 mmol/L and the urine ketones are not cleared yet, but a switch to a less intensive care is reasonable if the patient is stable and alert.

If the patient is a previous insulin treated dog, the pre-DKA insulin can be reinstituted, assuming the underlying precipitant factor has been controlled. For a newly diagnosed dog, we usually start at 0.5-1.0 U/kg BID of intermediated insulin, subcutaneously (higher dose for dogs with concurrent hyperadrenocorticism, for example). Regular insulin should be administered initially, and every 4-6 hours as needed, to maintain blood glucose concentrations between 100-250 mg/dL. Regular insulin takes 30 minutes to act, when given SC, and NPH insulin usually takes 2-3 hours. So, when making the transition, it is important to continue the IV insulin during the first 30 minutes after the SC injection of Regular insulin. If insulin interrupted, DKA may recur.

For cats we recommend begin with glargine and supplemental Regular insulin, even for those previously treated with diet only or combined with oral antihyperglycemic agents. The effect sulfonylurea-based therapies take weeks to begin. DKA takes hours.


1.  Duarte R, Simões DMN, Franchini ML, et al. Accuracy of serum β-hydroxybutyrate measurements for the diagnosis of diabetic ketoacidosis in 116 dogs. J Vet Intern Med 2002;16:411-417.

2.  Sheikh-Ali M, Karon BS, Basu A, et al. Can serum β-hydroxybutyrate be used to diagnose diabetic ketoacidosis? Diabetes Care 2008;31:643-647.

3.  Di Tommaso M, Aste G, Rocconi F, et al. Evaluation of an handheld electrochemical sensor for measuring whole blood 3-β-hydroxybutyrate concentration in diabetic dogs. Proc. 25th ACVIM Forum, Seattle, WA 2007:843.

4.  Roomp K, Rand JS, Planck M. Factors predictive of non-insulin dependence in diabetic cats initially treated with insulin. Proc. 26th ACVIM Forum, San Antonio, TX 2008: 862.

5.  Hume DZ, Drobatz KJ, Hess RS. Outcome of dogs with diabetic ketoacidosis: 127 dogs (1993-2003). J Vet Intern Med. 2006;20:547-555.

6.  Durocher LL, Hinchcliff KW, DiBartola SP, et al. Acid-base and hormonal abnormalities in dogs with naturally occurring diabetes mellitus. J Am Vet Med Assoc. 2008;232(9):1310-1320.


Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Ricardo Duarte, DVM, PhD
Hospital Veterinário Pompéia
São Paulo, SP, Brazil

MAIN : Endocrine : Diabetic Ketoacidosis
Powered By VIN