Treating a diabetic patient can be a very challenging task for the practitioner. In the 'perfect world', diabetic regulation would be black and white, but in the 'not-so-perfect world' several phenomena occur that can be confusing and potentially compromise the patient. The information that follows will explain most of these situations with the hopeful outcome being improved patient care.

Persistent Ketonuria Despite Clinical Improvement

This problem is caused by either an increased beta-hydroxybutyrate to acetoacetate ratio or high levels of acetone. These processes will reverse and ketonuria will disappear with continued treatment. Ketone production usually ceases after 3-4 days of treatment. The paradox is illustrated when a severely ill patient is admitted with negative ketones and then is ketones positive on Day 2 despite its clinical improvement. This occurs as a result of the excessive concentrations of betahydroxybutyrate being present on Day 1 now being metabolized through a reversal process by Day 2, allowing for the formation of acetoacetate which will go on to be further metabolized to bicarbonate and water.

Variables Affecting SQ Insulin

 Dose: the larger the dose, the longer the absorption time and duration of action.

 Injection site: mobile body parts (limbs) can allow for faster absorption.

 Depth of injection: IM route allows for faster absorption compared to SQ route. Remember, only regular crystalline insulin [R] can be given IM.

 Environmental temperature: insulin will be absorbed faster in warmer environment because of increased peripheral blood supply.

Exercise Associated Hypoglycemia

Exercised muscle is less insulin dependent for glucose transport. Therefore, the diabetic that is infrequently exposed to exercise will be insulin sensitive when that exercise occurs. This is prevented by giving less insulin or more energy substrate to the patient at the time exercise is undertaken.

Insulin Absorption Variability

This will be affected by insulin type, injection site, subcutaneous protease activity, and exercise. There are rare human diabetics who have increased subcutaneous tissue protease activity and therefore cannot benefit from SQ insulin injections. Alternative treatment routes are necessary in such patients. I have been suspicious for this when one canine patient required approximately 200 units of NPH insulin twice daily.

Early Morning Hyperglycemia

This can be caused by the waning effect of the previous insulin dose, the Somogyi phenomenon, and the dawn phenomenon. The transient effect of insulin can be corrected by using a longer acting insulin product such as, insulin glargine, or PZI. Dividing the dose in half and administering the insulin at 12-hour intervals is also very helpful.

The Somogyi effect is due to excessive insulin and requires a reduced dosage (25-50%) that is best given on a divided basis.

The dawn phenomenon occurs in humans and might be due to the transient effect of insulin or to the secretion of endogenous growth hormone that can antagonize the effect of insulin. This hypothesis has been challenged in humans because some of those affected people can have normal amounts of growth hormone.

Somogyi Reaction

This is characterized as posthypoglycemic hyperglycemia and is caused by excessive insulin dosage. The patient typically has an early hypoglycemic reaction followed by hyperglycemia the following morning. Factors playing a role in this are the insulin dosage and the patient's counter-regulatory hormones. The solution entails insulin dosage reduction by 25-50%. Splitting the dose into one half being given every 12 hours also helps.

Fever After Hypoglycemia

The usual reaction to hypoglycemia is hypothalamic-mediated hypothermia. Sometimes the hypothermia can stimulate excessive hypothalamosympathetic activity which causes increased calorigenesis and subsequently fever.

This problem usually self corrects. It is important not to let it prompt a search for FUO and unnecessary treatment.

Altered CNS Function in a Ketoacidotic and Hyperosmolar Diabetic

The main causes include paradoxical cerebrospinal fluid acidosis accompanying bicarbonate administration, altered CNS oxygenation due to disturbed O₂-Hb dissociation, and unfavorable osmotic gradients.

Cerebral Edema

Factors contributing to this include osmotic dysequilibrium, excessive hydration, hyponatremia, and sodium bicarbonate treatment. It is essential to avoid lowering the blood glucose level at rates exceeding 75-100 mg/dl/hr (4.1-5.0 mm/L/hr). Providing 0.9% NaCl will help maintain adequate serum sodium levels, and thus offset the formation of any osmotic gradients between the brain parenchyma and the extracellular fluid compartment which would cause an influx of water into the brain, resulting in cerebral edema.

Hyperosmolar Hyperglycemia Without Ketoacidosis

This is thought to occur when there are adequate amounts of insulin in the portal blood which can inhibit hepatic ketogenesis. However, there is not enough insulin produced to deter glucagon-induced gluconeogenesis and to allow for ample peripheral glucose utilization. The absence of serum ketones might also be the result of the majority of the organic ketone acids being betahydroxybutyrate instead of acetoacetate. The essential criteria for making this diagnosis is by demonstrating marked hyperglycemia with the absence of ketones, and renal insufficiency with azotemia and oliguria.

Progressive Hyperglycemia

The factors allowing for this are impaired peripheral glucose utilization, extracellular fluid volume depletion, and decreased urine output. With normal urine production the blood glucose concentration rarely exceeds 400-500 mg/dl (22.0-27.5 mm/L). Intravenous fluids alone will lower the blood glucose levels by up to 50% through enhanced renal glucose excretion, dilution of the blood glucose concentration, and allowing for decreased levels of counter-regulatory hormones.

Life-threatening Hypokalemia

Many factors can contribute to this problem including osmotic diuresis, reversal of metabolic acidosis, insulin administration, GI loss, and the clinician's failure to provide ample amounts. This can be managed by delaying insulin treatment for at least 4-12 hours while the potassium replacement ensues at a rate of 0.5 to 1.0 mEq/kg/hr (1 mEq/L = 1 mm/L). When the decline in serum potassium is life-threatening, the rate of potassium infusion can be increased to 1.5 mEq/L/hr with simultaneous ECG patient monitoring.

Reasons for Hyponatremia

Some are factitious (hypertriglyceridemia when sodium is measured with flame photometer, hyperglycemia) while others are real (osmotic diuresis, insulinopenia impairs renal tubular sodium absorption, GI loss).

Paradoxical CSF Acidosis

This is caused by excessive bicarbonate treatment which allows for decreased stimulation of the medulla respiratory center, allowing for increased PCO₂ production which can enter the CSF and further lower its pH by the carbonic anhydrase enzyme mechanism. Since there is a delay in bicarbonate absorption through the blood brain barrier, the formation of acid will occur before it can be offset by the buffering effect of bicarbonate. The result is a rapid occurrence of acidosis in the CSF and the possible demise of the patient. This is avoided by limiting bicarbonate treatment to only those patients with severe acidosis (blood pH < 7.0) and limiting the blood pH correction to 7.1-7.2.

Metabolic Alkalosis in DKA

This is associated with excessive vomiting allowing for the GI loss of H⁺ and Cl^- leaving the patient with an overall base excess. This will certainly worsen any existing hypokalemia. It is characterized as a low serum chloride level, low total CO₂ (bicarbonate), hypokalemia, and a blood pH that is unexpectedly higher than what would usually be seen with the low serum bicarbonate concentration.

Hypophosphatemia

This results from increased tissue catabolism, impaired tissue glucose utilization and cellular phosphorus uptakes, and increased renal excretion consequent to metabolic acidosis. The question of treating with potassium phosphate is controversial because clinically significant adverse effects are not as common as one is led to believe. Experimentally, the main concerns include hemolysis, rhabdomyolysis, impaired ATP production, and others. Correction can be made by using potassium phosphate solution as part of the potassium replacement therapy and by feeding the patient. Potassium phosphate salts must be given by slow IV injection in order to avoid causing hypocalcemia.

Diabetic Cataracts

Due to osmotic changes in the lens because of the activation of the intralenticular sorbitol-polyol pathway. This occurs as a result of saturation of the usual glucokinase enzyme reaction in the lens of the dog and the subsequent formation of sorbitol dehydrogenase and aldose reductase enzymes which promote the formation of fructose and sorbitol inside the lens. This sets up an osmotic gradient allowing for the influx of sodium ions and water which disrupts the lenticular architecture and cataract formation. This reaction rarely occurs in the cat thus explaining why cataracts are rare in diabetic cats.

Diabetic Neuropathy

This complication can cause substantial morbidity for the diabetic dog and cat. It is due to abnormal myelin formation because of several reasons, some of which include impaired blood supply to the nerve tissue, toxic radical formation as a result of pathologic oxidizing reactions, impaired myoinositol production within the nerve, glycosylation of basement membranes and other pathological biochemical reactions.