Diagnosing hyperadrenocorticism in the dog (Proceedings)

Pathophysiology

Hyperadrenocorticism (HAC) is a common endocrine disease of the dog. The disease and its clinical manifestations are either the result of excess cortisol production by the adrenal glands or iatrogenic due to administration of glucocorticoids. The normal pathophysiology of cortisol production involves the hypothalamus, pituitary gland and adrenal glands. The hypothalamus produces corticotropin releasing hormone (CRH) as a result of physical, emotional, chemical and other stressors. The CRH stimulates the pituitary gland to produce adrenocorticotropin hormone (ACTH). The ACTH then stimulates the adrenal glands to produce cortisol. Cortisol provides negative feedback to the hypothalamus and pituitary gland. ACTH also provides negative feedback to the pituitary. There are two forms of naturally occurring hyperadrenocorticism in the dog. The most common form is pituitary hyperadrenocorticism (PDH) and affects 80 to 85% of dogs with hyperadrenocorticism. Adrenal dependent or adrenocortical hyperadrenocorticism (ACH) is less common affecting the remaining 15 to 20% of dogs with HAC. There are rare reports of dogs with both pituitary and adrenal dependent HAC.

Pituitary dependent hyperadrenocorticism most commonly occurs as the result of a pituitary microadenoma (80 to 90%). Tumors are classified as microadenomas if they are less than 1 cm in diameter. Pituitary macroadenomas are less common (10 to 20%) and defined as tumors greater than 1 cm in size. This is clinically relevant because macroadenomas are much more likely than microadenomas to cause neurologic signs in addition to other signs of hyperadrenocorticism. Adenocarcinomas of the pituitary gland have also been reported but are extremely rare. Adenomas and adenocarcinomas result in increased ACTH production from the pituitary gland. Chronic increases in ACTH result in hyperplasia of the adrenal glands and increased cortisol secretion. A less common variant of bilateral hyperplasia is nodular hyperplasia. The nodules are believed to be secondary to PDH but can be confused with adrenal tumors on ultrasound or cross-sectional imaging.

Adrenocortical hyperadrenocorticism results from adenomas or adenocarcinomas affecting one (common) or both (rare) adrenal glands. When one adrenal is affected the other becomes atrophied and non-functional. Carcinomas and adenomas of the adrenal glands occur with about equal frequency. These tumors autonomously secrete excess amounts of cortisol independent of the influence of ACTH and, in fact, the excess cortisol causes decreases in CRH and ACTH.

Atypical hyperadrenocorticism occurs as a result of increased steroid hormones, or precursors to steroid hormones, other than cortisol which result in clinical signs consistent with hyperadrenocorticism. Tumors of the pituitary or adrenal gland may be involved. These hormones may be androstenedione, estradiol, progesterone, 17-OH progesterone, dehydroepiandrosterone sulfate and estradiol.

Signalment

This is typically a disease of middle-aged to older dogs but it can rarely occur in young dogs. Any breed can be affected but poodles, dachshunds, terriers and beagles have an increased incidence. Interestingly, approximately 75% of dogs with PDH are less than 20 kg. In dogs greater than 20 kg, there is equal frequency of PDH and ACH.

Clinical findings

It is important to remember that the diagnosis of HAC is dependent on clinical findings. Dogs with HAC may present with one or more of the following signs. The most common findings are polydipsia, polyuria, polyphagia, abdominal distention, panting, muscle wasting, cutaneous lesions and lethargy. Additional findings include lameness, hypertension, proteinuria, anestrus and testicular atrophy. In dogs with macroadenomas neurologic signs may be seen and include lethargy, dullness, confusion, pacing, ataxia, circling, blindness and seizures.

Laboratory tests

A minimum database is performed on all dogs suspected of HAC to identify findings supportive of HAC as well as to identify other potential causes of the clinical signs or concurrent disease. The CBC may be normal. A stress leukogram, increased red cells and thrombocytosis may also be noted.

Liver enzymes are elevated in the majority of dogs. Alkaline phosphatase (ALP) increases as a result of glycogen accumulation and hepatocellular vacuolation. There are several isoenzymes including a corticosteroid induced isoenzyme. Unfortunately this isoenzyme is not specific for excess endogenous or exogenous cortisol. Alanine transferase also increases secondary to damaged hepatocytes. The increase is typically not as marked as the ALP. Cholesterol is also commonly elevated due to the stimulation of lipolysis by cortisol. Lipemia is often present and suggestive of hypertriglyceridemia. Blood urea nitrogen and creatinine may be decreased secondary to polyuria. As mentioned before, some dogs with HAC also have diabetes mellitus so hyperglycemia might be noted. Typically these dogs present and are diagnosed with diabetes mellitus and it is only after establishing they have some form of insulin resistance that HAC is considered.

Urine may be variably concentrated. Dogs with HAC can concentrate their urine and if water is inadvertently withheld, or hospitalized dogs fail to drink, a normal urine specific gravity may be obtained. Glucosuria might result from hyperglycemia. Proteinuria may occur due to systemic hypertension or glomerular disease. A urine protein creatinine ratio is indicated to quantify significant proteinuria. Bacteria and white blood cells may or may not be noted in the presence of infection. This, in addition to the high incidence of infection in dogs with HAC, is reason to culture all newly diagnosed dogs.

Cortisol exerts a negative feedback on thyroid stimulating hormone which can result in secondary, but reversible, hypothyroidism. Glucocorticoids may also affect protein binding and peripheral metabolism of thyroid hormones.

Arterial blood gas analysis can be performed in dogs suspected of pulmonary thromboembolism. Hypoxemia, an increased alveolar gradient and hypocapnia may be noted. Hypoxemia may be due to ventilation/perfusion mismatch and intrapulmonary shunting. Hypocapnia is secondary to hyperventilation.

Radiography

Thoracic radiographs might reveal dystrophic mineralization. Mineralization may be seen as a generalized interstitial pattern, bronchial mineralization or tracheal mineralization. Dystrophic mineralization is a non-specific finding that can be seen in animals without HAC. Lesions consistent with pulmonary thromboembolism might be seen. This includes normal radiographs, aerated pulmonary parenchyma with decreased vasculature, alveolar disease, blunted pulmonary arteries, an enlarged main pulmonary artery, cardiomegaly and pleural effusion.

Abdominal radiographs will reveal hepatomegaly. Adrenal tumors can sometimes be seen and may be mineralized. A distended or incompletely emptied urinary bladder may be noted. Calcium-containing urinary calculi may be noted and are rare but occur with increased frequency in HAC.

Screening tests for hyperadrenocorticism

There is no one perfect test for diagnosing hyperadrenocorticism in dogs. No single test is 100% sensitive (picks up all dogs with the disease in question) and 100% specific (does not test positive individuals without the disease). It is important to combine clinical findings with diagnostic tests for this reason. These tests should be performed on animals suspected of having HAC based on historical, physical and initial laboratory diagnostics. The screening tests for hyperadrenocorticism are the ACTH stimulation test, low dose dexamethasone suppression (LDDS test) and urine cortisol creatinine ratio (UC:Cr). Basal cortisol levels are not reliable because of variation in cortisol levels at any given time.

ACTH stimulation test

The premise of the ACTH stimulation test is that administration of ACTH to a normal dog will stimulate the adrenals in a normal fashion but not excessively. ACTH gel (1 – 2.2 U/kg IM) or synthetic corsynotropin (5 mcg/kg IV or IM) is given. Heparinized plasma or serum is collected prior to and 2 hours after ACTH gel administration or prior to and 1 hour after administration of synthetic corsynotropin. In dogs with PDH the adrenals are hyperplastic and have an exaggerated response to additional ACTH. With ACH the autonomous cortisol-secreting cells of the adrenal tumor produce an exaggerated response as well. Because both forms of HAC produce an exaggerated response, this test can not be used to differentiate between pituitary and adrenal dependent hyperadrenocorticism. It is important to evaluate the absolute numbers, particularly the post stimulation value, and not the increase. Unfortunately the ACTH stimulation test might miss up to 40% of animals with HAC. If HAC is suspected in a dog with a normal response a LDDS test should be performed. If the results are suggestive of HAC then a differentiating test can be performed. Up to 15% of dogs with positive tests do not have HAC. The ACTH stimulation test is the only test available for monitoring therapy for HAC. This test is also able to identify dogs with iatrogenic HAC.

Low dose dexamethasone suppression test

The premise of the LDDS test is that when exogenous cortisol, even a small amount, is given to a normal dog the hypothalamus and pituitary will be suppressed so minimal ACTH and subsequent cortisol is produced. Cortisol will be suppressed within 3 hours and for up to 48 hours in normal dogs. Interestingly enough, in dogs with HAC circulating cortisol loses its effectiveness in 3 to 6 hours. Dexamethasone is given at 0.01 mg/kg IV and samples taken prior to, at 4 and 8 hours after administration. With HAC large amounts of cortisol are already being produced so the addition of a small amount of cortisol is ineffective in suppressing pituitary function. The 8 hour cortisol should be above the reference range with both forms of HAC. PDH can have several variations. With PDH the cortisol at 4 and 8 hours might be greater than baseline, same as baseline or has suppressed but remains greater than 50% of baseline. In another scenario the 4 hour cortisol might be less than the reference range or decrease to less than 50% of baseline but the 8 hour should again be above the reference range. It is also possible with PDH that the 8 hour cortisol is less than 50% of baseline but remains above the reference range. In approximately 40% of dogs with PDH and all dogs with ACH, there is no suppression of cortisol levels at any time and the 8 hour should be above the reference range. The sensitivity of this test for HAC is 85 to 95% thus 5 to 15% of dogs with HAC will test negative. Specificity is more variable and up to 25% of dogs without the disease will test positive. Reports in the literature are variable but in general the LDDS test is believed to have a slightly higher sensitivity (dogs with the disease will have a positive test) but poorer specificity (dogs without the disease may test positive) than the ACTH stimulation test.

Urine cortisol:Creatinine

The premise of the urine cortisol creatinine ratio is that animals chronically producing large amounts of cortisol will have increased amounts in their urine. Urine is collected at home to avoid stress that might affect cortisol levels. Normal dogs should have cortisol values within the reference range. Dogs with HAC should have levels above the reference range. This test is highly sensitive but poorly specific with reports of up to 80% of dogs with non adrenal illness testing positive. This test is used primarily to rule out HAC because a dog with a negative test is very unlikely to have HAC. I use this test when I have a low index of suspicion for HAC.

Steroid hormone profiles

Steroid hormone profiles are typically run in dogs suspected of HAC based on clinical signs that fail to exhibit increased cortisol levels on more traditional tests for HAC. It is possible that these animals are unable to produce excess cortisol due to a lack of enzymes necessary for its synthesis but some precursors might be found in excess. The protocol for the ACTH stimulation test is used but in addition to cortisol, 17-OH progesterone, estradiol, androstenedione, progesterone and aldosterone are measured. Dogs with atypical HAC will have increases in one or more of these 'atypical' steroid hormones. It is important to note that some of these other steroid hormones are affected by concurrent disease.

Differentiating tests for hyperadrenocorticism

Once the diagnosis of hyperadrenocorticism is confirmed PDH should be differentiated from ADH because of different therapeutic implications. Differentiation is done utilizing the LDDS test, high dose dexamethasone suppression (HDDS) test, endogenous ACTH, abdominal ultrasound and cross-sectional imaging.

Low dose dexamethasone suppression test

The LDDS test can also be used as a differentiating test. The protocol is as stated before. In the normal dog, dexamethasone persists for up to 48 hours but in dogs with PDH dexamethasone activity is only 3 to 6 hours. Approximately 60% of dogs with PDH will exhibit suppression at 4 hours but not at 8 hours. Since virtually no dogs with ACH suppress, this test can be used as a differentiating test if the patterns above are noted. If the dog fails to suppress at 4 and 8 hours then PDH or ACH may be present.

High dose dexamethasone suppression test

The high dose dexamethasone suppression test (HDDDS) differs from the LDDS test by the dose of dexamethasone. As discussed with the LDDS test, normal dogs suppress at 4 and 8 hours with dexamethasone. Higher doses of dexamethasone should suppress a pituitary tumor but not an autonomously secreting adrenal tumor. Dexamethasone is given at 0.1 mg/kg IV and pre, 4 hour and 8 hour cortisol levels are measured. Suppression occurs if the 8 hour is less than or equal to a 50% decrease from baseline or the 8 hour cortisol is within or below the reference range. The utility of the 4 hour cortisol is questionable since it is the 8 hour cortisol level that is used for interpretation. Dogs with PDH should adapt at least one of the criteria stated above. Unfortunately 25% of dogs with PDH fail to suppress as stated above. Almost all dogs with ACH fail to suppress.

Endogenous ACTH

Adrenocorticotropin hormone can be measured to help differentiate PDH from ADH. Blood is collected in an EDTA tube, the plasma spun off immediately and frozen until analysis. Endogenous ACTH should be low in dogs with ACH because large amounts of cortisol suppress ACTH release form the pituitary. With PDH endogenous ACTH would be increased since the pituitary tumor is responsible for excess ACTH production. Unfortunately many ACH and a few of the PDH have ACTH levels within the reference range and are considered non-diagnostic. The test could be performed at another date or other discriminatory tests done.

Abdominal ultrasound

Ultrasound is much more sensitive for detecting adrenal masses than radiography. Normal adrenal glands are reportedly less than 7.4 mm diameter at the caudal pole of the left adrenal. Ultrasound can be used to help differentiate PDH from ACH but adrenal masses found may not be functional cortisol-secreting tumors. In a dog with the appropriate clinical findings, two normal or bilaterally enlarged adrenals, PDH is likely. Unilateral adrenomegaly, especially in the face of a small or undetectable contralateral adrenal gland, is suggestive of ACH. Larger, irregular adrenal masses that invade surrounding structures are more suggestive of malignant adrenal tumors. The mass may be mineralized (shadow) or associated with a thrombus. The liver can also be assessed for evidence of metastasis in the face of an adrenal tumor.

Cross-sectional imaging

Cross-sectional imaging can be used as a differentiating test as well. CT and MRI allow you to visualize the pituitary gland and measure the size of the mass. This can be important for treatment recommendations because irradiation of the pituitary might be recommended for larger tumors. Approximately 60% of dogs with PDH have masses visible with CT or MRI. CT or MRI can also be done with ADH to evaluate the adrenal for size, local invasion and metastasis.

References

Auriemma E. Barthez PY, van der Vlugt-meijer RH, et al. Computed tomography and low-filed magnetic resonance imaging of the pituitary gland in dogs with pituitary-dependent hyperadrenocorticism: 11 cases (2001-2003). J Am Vet Med Assoc 2009 Aug 15;235(4):409-14.

Benitah N, Feldman EC, et al. Evaluation of 17-hydroxyprogesterone concentration after administration of ACTH in dogs with hyperadrenocorticism. J Am Vet Med Assoc. 2005 Oct 1;227(7):1095-101.

Bertoy EH, Feldman EC, Nelson RW, et al. One-year follow-up evaluation of magnetic resonance imaging of the brain in dogs with pituitary-dependent hyperadrenocorticism. J Am Vet Med Assoc 1996 Apr 15; 208(8):1268-73.

Chapman PS, Mooney CT, et al. Evaluation of the basal and post-adrenocorticotrophic hormone serum concentrations of 17-hydroxyprogesterone for the diagnosis of hyperadrenocorticism in dogs. Vet Rec. 2003 Dec 20-27;153(25):771-5.

Ettinger SJ, Feldman EC. Textbook of Veterinary Internal Medicine. 6th ed. Elsevier. Pp 1592 – 1611.

Feldman EC, Nelson RW Canine and Feline Endocrinology and Reproduction. 3rd ed. Saunders. Pp 251-353.

Hill KE, Scott-Montcrieff JC, et al. Secretion of sex hormones in dogs with adrenal dysfunction. J Am Vet med Assoc. 2005 Feb 15;226(4)556-61.

Oliver JW. Steroid Profiles in the diagnosis of canine adrenal disorders. Proceedings 25 th ACVIM 471-3, 2007.

Nyland TG, Mattoon JS, Herrgesell EJ, et al. Adrenal glands in Small Animal Diagnostic Ultrasound. Second ed. Nyland TG, Mattoon JS ed 2002. pp196-206.

Penninck DG, Feldman EC, Nyland TG. Radiographic features of canine hyperadrenocorticism caused by autonomously functioning adrenocortical tumors: 23 tumors (1978-1986). J Am Vet Med Assoc 1988 Jun 1;192(11):1604-8.

Ristic JM, Ramsey IK, et al. The use of 17-hydroxyprogesterone in the diagnosis of canine hyperadrenocorticism, J Vet Intern Med. 2002 Jul-Aug;16(4):433-9.

Syme HM, Scott-Montcrieff JC, et al. Hyperadrenocorticism associated with excessive sex hormone production by an adrenocortical tumor in two dogs. J Am Vet med Assoc. 2001 Dec 15;219(2):1725-8,1707-8.