Diagnosis and management of the cushingoid diabetic dog (Proceedings)

Diabetes Mellitus (DM) and hyperadrenocorticism (HAC) are common endocrinopathies in dogs that often occur simultaneously in the same patient. Diagnosis and management of concurrent disease may be a challenge to the practitioner since many clinical signs of DM and HAC are similar.  In animals with concurrent disease, the practitioner should first get some control of the DM and diagnose the HAC later.  Although the clinician may want to treat the HAC first in order to facilitate better control of the DM, diagnostic tests available for HAC are less reliable in poorly regulated diabetic patients.

Treatment of diabetes mellitus

The initial treatment of DM in the dog includes a long-acting insulin (NPH, Vetsulin (if available), PZI (ProZincTM), Glargine) at an initial dose of 0.5 U/kg given subcutaneously twice per day after feeding. The dog is re-examined 10-14 days later with a potential glucose curve to determine if changes in the dose and type of insulin need to be made.  The owners are asked to monitor appetite, water consumption, body weight, and attitude at home.  They are also given urine dipsticks to record glucosuria and ketonuria.  Alternatively, the owners may measure blood glucose via a portable glucometer, although ketosis will not be assessed.  The owners should be instructed not to change the dose of insulin based upon urine/blood glucose measurements without the input of the veterinarian, and the insulin dose should remain constant for the initial 10-14 days of treatment so that the dog can fully metabolize the insulin. Glucose monitoring is performed every 10-14 days and insulin dose adjusted accordingly until an appropriate dose is reached.  In most cases, adequate control of DM (consistent blood glucose concentration between 100-250 mg/dl) is achieved within 4-6 weeks.  Insulin resistance is suspected when glycemic control is not achieved, and the dose of insulin is above 1.5U/kg per injection. If insulin resistance is encountered concurrent disease, especially urinary tract infections, must be investigated.  Insulin resistance may signal the presence of concurrent Cushing's disease.

Diagnosis of hyperadrenocorticism in patients with diabetes mellitus

Adrenal function testing of a patient with DM is indicated when insulin resistance is suspected and/or clinical and clinicopathologic changes suggestive of HAC persist with appropriate insulin therapy.   Screening tests for HAC are not performed until some glycemic control (blood glucose 100-350 mg/dl) is achieved with insulin since patients that are extremely hypo- or hyperglycemic may have falsely positive test results. The screening tests most commonly used to diagnose HAC in the diabetic patient are the low dose dexamethasone suppression (LDDS) test and the ACTH stimulation test.  Of these tests, the LDDS has the greatest sensitivity (likelihood that a negative test result is found in animals without the disease) of approximately 90%.  However, the specificity (likelihood that a positive test result indicates the presence of the disease) of the LDDS is 70-73% and is least specific in animals with concurrent disease, especially diabetes mellitus.  The ACTH stimulation has a sensitivity of approximately 80%; however the specificity is greater than that of the LDDS in dogs with concurrent disease.  Therefore, in dogs with DM that are suspected of having HAC, the ACTH stimulation test is the superior screening test.  A urine cortisol:creatinine ratio is not recommended as a screening test for HAC in dogs with DM because the ratio is usually elevated due to the stress of DM alone.

Diabetic dogs are fed and given their insulin dose as usual prior to adrenal function testing.  The dog is admitted to the hospital for an ACTH stimulation test and a simultaneous glucose curve.  It is helpful to obtain at least one glucose measurement prior to performing the ACTH stimulation test to get an idea of the control of the DM.  The blood glucose measurement should be between 100-350 mg/dl.  If this is not the case, the ACTH stimulation test may not be accurate and should be repeated when blood glucose control is better. The ACTH stimulation test is performed by measuring plasma or serum (depending upon the laboratory used) cortisol concentration prior to and 45-60 minutes following the IV injection of 5 mcg/kg of synthetic ACTH.  A post stimulation cortisol concentration less than 18 ug/dl does not support a diagnosis of HAC; a post stimulation cortisol concentration above 22 ug/dl indicates the presence of HAC.  If the test results are not indicative of HAC, but the practitioner has strong clinical suspicion of the disease, the ACTH stimulation test may be repeated in 2-4 weeks.  The practitioner must be aware that a falsely positive ACTH stimulation test result may occur.

Although less specific than the ACTH stimulation test in dogs with concurrent DM, the LDDS may be performed.  A simultaneous glucose curve should be performed during the test to ensure that the blood glucose concentration is between 100-350 mg/dl.  If this is not the case, the test should be aborted.  The LDDS test is performed by administering 0.01 mg/kg of dexamethasone intravenously (IV) and measuring plasma or serum cortisol concentration before and four and eight hours after the injection.  In normal dogs, the cortisol concentration should be suppressed below the normal level for 8 hours after injection of dexamethasone.  In dogs with HAC the 8 hour post injection cortisol  concentration is elevated (>1.4 mg/dl).   Although a false positive LDDS test is possible, adrenal function test results have been shown to be normal in colony dogs with chronic, well-regulated diabetes mellitus and normal adrenal function.  

Differentiating pituitary dependent versus adrenal HAC is indicated only if the results of the LDDS test or ACTH stimulation test are diagnostic of HAC.  A LDDS test may also serve as a differentiating test in some instances.  A LDDS test is diagnostic of PDH if the 4 hour cortisol concentration is <1.4 mg/dl indicating initial suppression, and the 8 hour cortisol concentration is >1.4 mg/dl.  A LDDS test is also diagnostic of PDH if the 4 or 8 hour cortisol concentration is less than 50% of the baseline cortisol concentration.   Other differentiating tests are the measurement of endogenous ACTH concentration and the use of abdominal ultrasound to detect adrenal tumors.  Endogenous ACTH concentration measures the basal level of ACTH.  In dogs with adrenal tumors, the ACTH level will be low.  In dogs with pituitary-dependent disease, the ACTH level will be normal or high.  Abdominal ultrasound can be used to measure adrenal size.  If an adrenal tumor is present, the adrenal affected will usually be enlarged and the other adrenal small due to atrophy in the presence of elevated blood cortisol levels.

Initial treatment of hyperadrenocorticism in a patient with diabetes mellitus

The most common form of HAC is pituitary-dependent which may be treated effectively with o,p'-DDD (Lysodren) or trilostane.   The treatment of PDH with o,p'-DDD is divided into an induction phase and a maintenance phase.  During the induction phase o,p'-DDD is given at a dose of 25-50 mg/kg PO, divided twice a day, for 7 to 10 consecutive days.   Smaller dogs require the higher end of this dose range whereas larger dogs require less.  The dose is ultimately influenced by the size of the tablets since o,p'-DDD is available in 500 mg tablets and cannot be reliably broken into pieces smaller than quarter tablets.  Mitotane can be compounded, but this raises the price substantially.  Trilostane may be started at 3 mg/kg twice per day.

Adrenal tumors are treated with surgery, o,p'-DDD, trilostane or combinations of the three.  Medical treatment of adrenal tumors with o,p'-DDD generally requires a higher induction dose, a longer induction period, and a higher maintenance dose.  Trilostane treatment can be started with standard doses.

In dogs with concurrent HAC and diabetes mellitus o,p'-DDD induction or initiation with trilostane treatment is performed cautiously, and the lower end of the o,p'-DDD  and trilostane dose is recommended.  Treatment of HAC will decrease endogenous glucocorticoids and their antagonistic effect on insulin.  As a result, the insulin administered will have increased potency, and doses that previously resulted in hyperglycemia may now cause hypoglycemia.  During the induction phase with o,p'-DDD or trilostane, the total dose of insulin is decreased by 10- 25%, depending on the size of the dog and the current dose of insulin.  The dose of insulin is decreased in an attempt to create mild hyperglycemia (150-350 mg/dl) during initiation of treatment for HAC.  The owner is instructed to sample the urine frequently (3 times a day).  If glucosuria is not detected 3 or 4 times in a row, the owner is asked to stop the o,p'-DDD or trilostane treatment and decrease the dose of insulin by half. Alternatively, the owner may measure blood glucose at home several times daily.  If the animal has a blood glucose less than 100, the veterinarian should be contacted and the animal should be given ½ of its normal insulin dose. An ACTH stimulation test and glucose curve are performed immediately to determine adequate o,p'-DDD or trilostane and insulin doses.  Physiological doses of prednisone (0.2 mg/kg PO, once a day) may be administered during the induction phase.  The low dose of prednisone is not likely to cause significant metabolic or clinical changes in view of the chronically elevated glucocorticoids to which the patient has been exposed, and it may help prevent hypoadrenocortical or hypoglycemic crises.  If the dog is not supplemented with a physiological dose of prednisone, prednisone and instructions to administer the medication if necessary are given to the owner.  The owner should be instructed to monitor the dog for lethargy, anorexia, vomiting, ataxia, or seizure activity, and induction should occur during a time when the owner can adequately supervise the patient in order that these parameters can be closely watched.  These signs may signal a hypoadrenocortical or hypoglycemic crisis.  In such an event, the owner should treat the animal with prednisone and Karo syrup and to seek emergency veterinary care.

Treatment and monitoring of concurrent hyperadrenocorticism and diabetes mellitus

In order to monitor o,p'-DDD and trilostane induction, an ACTH stimulation test is performed 24 hours after the last induction dose of o,p'-DDD was given and 48 hours after the last dose of prednisone (if used) was administered. The ACTH stimulation test should be performed 4-6 hours post trilostane treatment 10-14 days after the initiation of trilostane.  Sodium and potassium should also be measured in the serum in trilostane treated animals to assess mineralocorticoid activity.  The ACTH stimulation test is performed to determine the glucocorticoid reserve capacity of the adrenal glands.  Although the purpose of o,p'-DDD or trilostane treatment is to decrease glucocorticoid secretion, some reserve capacity is desired.  The owner is instructed to feed the dog and administer the usual dose of insulin on the day of the ACTH stimulation test.  A glucose curve is also performed that day.

Maintenance therapy with o,p'-DDD or trilostane  is indicated if cortisol concentrations before and after the ACTH stimulation are between 1-4mg/dl.   When cortisol concentrations before and after the ACTH stimulation are below 1mg/dl the dog is considered glucocorticoid deficient.  In this case, the o,p'-DDD or trilostane is discontinued, and sodium and potassium concentrations are measured to assess mineralocorticoid function.  If only glucocorticoid secretion is affected, as is the case with most dogs, prednisone supplementation alone is indicated.   In the rare instances in which electrolyte abnormalities are evident, supplementation with a mineralocorticoid is warranted, as well.   In cases of glucocorticoid and/or mineralocorticoid deficiencies, an ACTH stimulation test and electrolyte measurements are repeated in one week and monitored thereafter.  Treatment with o,p'-DDD or trilostane is discontinued until the pre-and post-ACTH stimulation cortisol concentrations are above 1 mg/dl.  o,p'-DDD or trilostane treatment is also discontinued, and an ACTH stimulation test is performed if the dog develops signs suggestive of hypoadrenocorticism such as lethargy, weakness, anorexia, vomiting, diarrhea, shaking, or shivering.

When interpreting the results of an ACTH stimulation test it is important to remember that numerical guidelines are a general recommendation that must be evaluated and sometimes altered in view of the clinical assessment.  The clinical signs are an important tool in assessing response to therapy.  Additionally, there is individual variation in response to therapy, and with time the optimal values for each patient become familiar.  Furthermore, when test results do not reflect the clinical presentation, they should always be questioned and repeated.

The results of the glucose curve are interpreted in light of the ACTH stimulation test.  If the ACTH stimulation test shows that the HAC is well regulated, or that the dog is glucocorticoid deficient, tight glycemic control (blood glucose 100-250 mg/dl) is attempted, and the dose of insulin is adjusted accordingly.  However, if the HAC is not well regulated, and o,p'-DDD or trilostane induction is continued, the dog is maintained at a mildly hyperglycemic state (blood glucose 150-350 mg/dl) in anticipation of a further decrease in insulin resistance.  Maintenance therapy with o,p'-DDD or trilostane begins once the ACTH stimulation test results show that the HAC is well controlled.  Tight glucose regulation with insulin can finally be attempted.  A glucose curve is performed, and an adequate dose of insulin is identified. 

The importance of routine urinalysis, urine sediment, and urine culture and sensitivity cannot be overemphasized.  Urinary tract infections may be difficult to diagnose in animals with either diabetes mellitus or HAC and can be very challenging to diagnose in dogs with both diseases.  Repeated urine cultures, whenever clinically indicated, may be necessary for the diagnosis.  Such infections must not be overlooked because they may result in sepsis in these immunocompromized animals.