Podcasts

Join Now

Risk factors and outcome predictors in cats with diabetic ketoacidosis (DKA) | VETgirl Veterinary Continuing Education Podcasts

Need Help?

In today’s VETgirl online veterinary continuing education podcast, we discuss diabetic ketoacidosis (DKA). Do you see cats with DKA in your practice? Does DKA really stand for “Diabetes Kills Animals?” (No.) What do you usually tell owners about their prognosis and what do you base that information on? While we know that DKA is a complicated form of diabetes mellitus (DM) and we often tell owners that it usually happens due to some other complicating concurrent disease or condition, we don’t actually know a lot about those diseases or conditions, and whether or not they contribute to the patient’s outcome.

So, Cooper et al out of University of Pennsylvania School of Veterinary Medicine wanted to try to answer that question, so they looked at cats with DKA, DM and no diabetic disease over a 10 year period. In this retrospective study, the authors compared variables to see if anything (such as signalment, concurrent conditions, etc.) seemed to affect outcome. Inclusion criteria for this study included cats with DKA that were hyperglycemic (>250 mg/dL), had a venous pH of <7.35, and that had evidence of ketonuria or ketosis (urine or serum acetoacetate >1.5 mmol/L). For cats with DM, inclusion criteria included persistent hyperglycemia (>250 mg/dL), glucosuria, no evidence of ketonuria and a venous pH ≥7.35. The non-diabetic disease cats were patients examined for preventative health care and routine wellness visits over the same time period. In the DKA group, treatment included a continuous infusion of regular (e.g., Humulin-R) insulin, added to 240 ml of 0.9% NaCl at either 1.1 U/kg or 2.2 U/kg body weight. The solution was adjusted based on blood glucose concentrations over the course of treatment.

Overall, 93 cats were included in the DKA group, 682 in the DM group, and 16,926 in the nondiabetic disease group. When the authors compared these groups, they found significant differences between the numbers of Abyssinian cats that had DKA or DM and those that were nondiabetic, and between the numbers of Siamese cats that had DKA and those that had DM or were nondiabetic. Male cats were more likely to have DKA or DM as compared to female cats. (Don’t buy a male Abby or Siamese). Of the cats with DKA, 50% (47/93) were diagnosed with DM concurrently, and 50% (46/93) had been previously diagnosed and were being treated for DM. Of those that were previously diagnosed, the median time between diagnosis of DM and DKA was 1 week (range 0.1-378 weeks).

Of the physical exam findings in the cats with DKA, 62% (58/93) were noted to have neurologic abnormalities, including 41 with dull mentation and 13 with obtundation. 24 cats were recumbent on presentation. As for clinicopathologic data, CBC results were available for 76 cats with DKA. Neutrophilia and thrombocytopenia were not associated with a diagnosis of acute pancreatitis, nor was neutrophilia associated with a urinary tract infection. Anemia and serum phosphorus concentrations were not correlated. Serum chemistry results were also available for 76 DKA cats. Hypokalemia (58%), hypophosphatemia (33%) and hypomagnesemia (14%) were common on presentation, and were documented in 97%, 65% and 76% of cats at least once during hospitalization with median times to the lowest potassium, phosphorus and magnesium concentrations being 17, 23.5 and 35 hours from the time of initial examination.
On presentation, median pH in cats with DKA was 7.19, and pH decreased in 66% (55.84) of cats during hospitalization, with the median time to lowest pH at 8 hours following initial exam. Low venous pH was inversely associated with increased ionized magnesium and increased BUN concentration (p=0.02, 0.01 respectively).

Of the cats with DKA, 73 had aerobic urine culture performed; 10 cats (14%) had bacterial growth and 63 (86%) did not. E. coli was the most commonly isolated organism (8/10 cats). Concurrent conditions in the cats with DKA included acute pancreatitis (34%), hepatic lipidosis (13%) and chronic kidney disease (13%). Overall, 51/93 (55%) cats with DKA had a concurrent condition including those just mentioned or a urinary tract infection, and some cats had more than one condition.

In general, treatment of cats with DKA included fluid therapy, potassium and other electrolyte supplementation, sodium bicarbonate, and insulin. Cats treated with a higher concentration of insulin via CRI (2.2 U/kg/240 ml bag of 0.9% NaCl versus 1.1 U/kg) were less likely to have a poor outcome (p=0.02).
50/93 (54%) cats were discharged home following treatment for DKA and were considered to have a good outcome. Of the remaining 43 cats, 33 were euthanized, 4 were transferred to the rDVM for further care, 3 were taken home against medical advice, and 3 cats died in the hospital. Increased initial creatinine, BUN, total serum magnesium and total bilirubin were associated with poor outcomes (p=0.005-0.03). No other factors were associated with outcome.

So, what can we take away from this VETgirl podcast?

Siamese cats may be at increased risk for DKA and Abyssinian cats may be at increased risk for DM and DKA when compared to cats with diabetic disease. While previous studies have suggested that Burmese cats in Australia and the UK and purebred cats in the US may be at increased risk for DM, this is the first study to suggest that these breeds specifically may be at risk.

The only factors associated with poor outcomes in cats with DKA in this study were increased initial creatinine, BUN, total magnesium and total bilirubin. Because increased creatinine and total magnesium were positively correlated with one another, the authors suggested that these values indicated renal dysfunction, and that the presence of renal dysfunction might be predictive of a poor outcome in cats with DKA. However, they did not find a correlation between the presence of chronic kidney disease and outcome, possibly because they defined it as azotemia after at least 72 hours of fluid therapy, and many cats had already been euthanized by that time.

Total bilirubin concentration was associated with poor outcome as well, although patients with conditions that would likely cause an increase in total bilirubin (e.g., acute pancreatitis, hepatic lipidosis) were not more likely to have a worse outcome. Many more cats had blood work performed to measure total bilirubin than further diagnostics performed to look for causes of increases in total bilirubin, so further research is needed in this area to determine why this association seems to exist.

Concurrent conditions, with acute pancreatitis being the most common, were diagnosed in over 50% of cats with DKA in this study. As we generally counsel owners, this study affirms that it is important to look for underlying diseases or conditions that might contribute to illness when a cat presents with DKA.
Interestingly, this study suggests that an insulin infusion at a higher concentration of 2.2 U/kg/240 ml bag of 0.9% NaCl may be preferred as it was associated with better outcomes. The authors hypothesize that this might be the case because the higher dose of insulin might just be more effective (it’s closer to the human and canine standard doses) or it might be due to other effects of insulin such as its anti-inflammatory effects. Also, most cats developed electrolyte abnormalities within the first 35 hours of treatment, highlighting the importance of frequent monitoring.

The pros of this study? Despite it being a retrospective, this is a fairly large study that looked at pretty big control populations, which is great. It would have been nice to see more comparisons between the control populations and the DKA cats; the authors really looked only at signalment for comparison. This study confirms, as we often suspect, that a large proportion of cats with DKA have concurrent conditions, so it is a nice jumping-off point for further studies looking at these conditions. It would also be interesting to look at variable insulin infusion concentrations prospectively.

In conclusion, this retrospective study suggests that a large proportion of cats with DKA have concurrent conditions, most commonly acute pancreatitis. Patients with DKA have a guarded prognosis (39% had a poor outcome in this study), so looking for underlying conditions that may be treatable, as well as measuring renal values, total magnesium and total bilirubin, may be helpful on presentation to help prognosticate for the pet owner. When in doubt, use a higher dose insulin infusion in your cats with DKA, as it may contribute to improved outcomes.

References:
Cooper RL, Drobatz KJ, Lennon EM, Hess RS. Retrospective evaluation of risk factors and outcome predictors in cats with diabetic ketoacidosis (1997-2007): 93 cases. J Vet Emerg Crit Care 2015;25(2):263-272.

Only VETgirl members can leave comments. Sign In or Join VETgirl now!