Treating the Complicated Diabetic Patient
World Small Animal Veterinary Association World Congress Proceedings, 2001
David Church

What is the difference between two apparently similar diabetic patients, one of which is superbly stabilised on once daily insulin, and the other which never seems to have a single day of adequate control no matter what insulin regime is implemented? There are many factors which may contribute to so called “brittle” diabetes however it is worth remembering that, at present, many remain unexplained and it is up to the attending veterinarian, often in conjunction with a specialist, to devise a management strategy which will maximise control and, in particular, minimise the risk of clinically significant hypoglycemia.


There is an extensive list of possible causes for a diabetic dog or cat to be difficult to stabilise or to keep well controlled. Perhaps concurrent possibly otherwise occult, hyperadrenocorticism must always be considered and ruled out.

Concurrent infections, progestin therapy, dysmenorrhea, acromegaly, thyroid dysfunction and general non-specific concurrent neoplasia must all be considered and evaluated as they may not produce otherwise obvious clinical signs but have profound effects on diabetic control. Any process that results in significant renal or hepatic dysfunction can make managing a diabetic a particularly difficult task.

Patients with significant pancreatic inflammation can present both a diagnostic and therapeutic challenge. Particularly sub-acute pancreatitis can produce repeated relapses in the general health of the animal along with altered insulin requirements. This presents little problem to the acute clinician when the animal demonstrates the “classical” signs of pancreatitis, however many diabetic patients with “smouldering” pancreatic inflammation may simply present “unwell” with no biochemical evidence to suggest active or ongoing pancreatic destruction. This is particularly true in cats where the situation is further clouded by the even less reliable association between hyperamylasemia/hyperlipasemia and active pancreatic inflammation.

In any patient in whom some form of pancreatitis is suspected as a possible complicating factor in diabetic management, specialist abdominal ultrasonography should be considered as one more mechanism by which this process can be clarified.

However, always remember there are a number of diabetic patients, both dogs and cats that are “idiopathically brittle”. These patients have seemingly no underlying disease and yet can prove remarkably difficult to stabilise. Usually these patients are difficult to stabilise from the outset. In those patients who have been well controlled for some time and suddenly drop out of control, the reason is usually the development of some concurrent disease.


Generally, whenever possible, correcting the underlying cause is the obvious solution although this is frequently difficult to achieve. Whenever possible, using smaller doses of rapidly acting insulin that coincide with meals, is also an excellent means of improving control, however for this to be feasible this practice has to be compatible with the household's normal activities.


Usually a diabetic patient that is unwell, either because of intercurrent disease or due to the endogenous toxemia that results from marked ketosis, will be inappetent, making standard subcutaneous insulin therapy difficult. In these cases, continuous intravenous insulin can be a very effective means of short-term management. Furthermore, as long as there is a volumetric pump available, a continuous intravenous insulin infusion is the simplest and least labor-intensive means for treatment of these patients.


These patients usually require total parenteral hydration, and by virtue of their poorly controlled diabetes, have relatively high fluid maintenance requirements. Consequently, flowing intravenous fluids at around 150 ml/kg/24hr generally will provide some replacement and adequate maintenance. The optimum fluid composition is 0.45% NaCl with 30–40 mmol/l of KCl added. It is often prudent to add the potassium as equal amounts of potassium chloride and potassium phosphate. This not only minimizes the risk of hypokalemia, but also of hypophosphatemia. While one aim is to rehydrate the patient, the other must be to provide some measure of diabetic control or at least inhibit ongoing peripheral lipolysis and hence to start to reduce the potential for ketoacidosis.


When insulin therapy is being considered to treat clinically significant ketoacidosis, the choice is continuous intravenous insulin therapy or repetitive intramuscular insulin injections. In both situations, the insulin should be in a soluble, relatively rapidly acting form. Although an intravenous insulin infusion sounds daunting it, is certainly the simplest and least labor intensive means of treating diabetic patients with GIT signs sufficiently severe to warrant a “nil per os” recommendation. In this situation, we are trying to adequate glucose homeostasis in a patient were parenteral drug and nutrient delivery is the only available route for administration.

One easy method for setting up the intravenous insulin fluid line is to add 25 units of soluble insulin to a 500 ml bag of fluids, producing an insulin concentration of 50 mU/ml. Since the standard insulin infusion rate is 40–60 mU/kg/hr, a convenient compromise is to flow 1 ml/kg/hr of this solution. Obviously, a flow rate of approximately 1 ml/kg/hr is inadequate for maintenance. Consequently, intravenous insulin has to be administered through a second infusion line, usually attached to the “Y” piece of the maintenance fluid line.

The insulin is infused at a rate of 1 ml/kg/hr until plasma glucose concentrations fall to 10–12 mmol/l. At this time the flow rate should be halved (0.5 ml/kg/hr) and a concurrent dextrose infusion introduced through the maintenance fluid line. One simple and effective means of achieving a balance between the insulin and glucose infused is to change the maintenance fluids from 0.45% NaCl and 30 mmol/l K to 0.45% NaCl and 2.5% dextrose with 30 mmol/l KCl while continuing to run this combination at 150 ml/kg/hr. This will introduce a glucose infusion rate of around 150 mg/kg/hr that should balance the insulin being infused at 0.5 ml/kg/hr.

During this time, the patient's blood glucose and plasma potassium are checked regularly. Over a period of 48–72 hours, the blood glucose should remain relatively steady, ketonemia, when present, should disappear, and generally, the patients return to a normal water and nutrient intake. Once this level of normality is established, it is likely that, at least in the short term, the patient will be able to be stabilised on a regular feeding regime and a regular insulin dose regime.

For a variety of reasons, it will not always be possible to administer an intravenous insulin infusion to an anorectic diabetic patient who requires relatively urgent insulin therapy. On some occasions, the clinician may try and use lower doses of a subcutaneously administered lente preparation twice daily. Although this practice may be effective and relatively safe if the dose is kept at around 0.25U/kg/12hr, there is always the possibility of giving a little too much. Any overdose in this situation is likely to last at least 12 hours.

Consequently, in the absence of a reliable infusion pump, a more attractive alternative to a continuous intravenous insulin infusion is repeated intramuscular injections of soluble insulin.

Dogs and cats should receive soluble insulin at a loading dose of 0.2 U/kg followed by 0.1 U/kg hourly thereafter. All injections should be administered intramuscularly. Blood glucose levels are monitored regularly (every 2–3 hours) until they have fallen to 12 -15 mmol/l. At this point, it is usually satisfactory to change to subcutaneous soluble insulin every six hours (0.2–0.4 U/kg). Because of the likely persistence of ketonemia and hence inappetence, it is important to maintain adequate blood glucose concentrations by introducing dextrose into the intravenous fluids. Generally, a 2.5% dextrose solution run at twice normal maintenance rates is satisfactory, although blood glucose levels should be checked every four to six hours. 

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

David Church

MAIN : : Complicated Diabetic Patient
Powered By VIN