Cushing’s syndrome (CS) is the term used to describe a range of diseases that result from long-term exposure to glucocorticoids. The most important glucocorticoid is hydrocortisone, a.k.a. cortisol. There are many widely used synthetic glucocorticoids (prednisone, prednisolone, methylprednisolone, triamcinolone, betamethasone, dexamethasone, etc.) and chronic exposure to any of these results in CS.

Naturally occurring CS is most often secondary to an ACTH secreting pituitary tumor -or- due to unregulated secretion of cortisol from hyperplasia or neoplasia of the adrenal cortex.¹ Very rarely, ACTH or cortisol might be secreted in excess from elsewhere in the body.

Veterinary endocrinologists have long professed that the clinical suspicion for CS should be based on supportive historical and physical examination findings.² Otherwise stated CS should be an “exam room diagnosis.”³

As our understanding of CS has grown and the availability of lab testing increased, a more recent classification of “Subdiagnostic Cushing’s Syndrome” has emerged to help explain situations where the dog appears to have CS however the ACTH stimulation test and the Low Dose Dexamethasone Suppression test (LDDST) are normal.¹ In these cases, hypercortisolemia simply cannot be proven despite the highest clinical suspicion, or, an adrenal tumor is secreting non-cortisol glucocorticoid(s), previously referred to as “atypical” Cushing’s Syndrome.

These changing definitions demonstrate that Cushing’s really is a variety of possible hormone imbalances resulting in a spectrum of clinical presentations.

Diagnosis

What is Your Clinical Suspicion?

A high clinical suspicion indicates that you believe there is a high chance of the diagnosis. Importantly, the likelihood that a specific patient has the disease in question (i.e., the pre-test probability) impacts the “trustworthiness” of the test results. It is the goal of the practitioner to evaluate the data, select the appropriate tests, and accurately diagnose the patient.

Remember that the signalment, history, and physical examination findings are some of the most important (and free) pieces of data available.

Signalment

Spontaneous Cushing’s Syndrome is typically seen in dogs around 10 years of age and is unlikely in dogs <4-years old. Three of every 4 dogs with CS are (or should weigh) less than 20 kg lean body weight.⁴ Large-breed dogs are over-represented in the roughly 20% of total cases caused by an adrenal tumor, however, in general, CS is uncommon in the large breed dog.

History

Dogs with CS rarely feel “sick” and should not have signs of illness (anorexia, vomiting, diarrhea, sneezing/coughing) at the time of workup.

The most common clinical signs of CS include the five P’s—polyuria (PU), polydipsia (PD), polyphagia (PP), panting, and pot-bellied appearance (hepatomegaly and weakened abdominal girdle). Additional clinical signs include skin and hair-coat changes (alopecia, thin skin, comedones, calcinosis cutis) as well as broader symptoms of muscle wasting and weakness (difficulty ambulating, difficulty jumping in the car, “splaying” out on slick surfaces) and a predisposition to recurrent infections. Insulin resistance and diabetes mellitus were noted in about 1 in 7 dogs with CS in one population.⁵

Remember that clinical signs are slow in onset and progression. Dogs rarely all have all the clinical signs at once and the clinical signs could be mild. Some dogs present with only skin and haircoat changes.

Minimum Database (CBC/Chemistry/UA)

Cushing’s should rarely be pursued because of the results of routine serum biochemistry; however, the routine minimum database is an important screening test for CS.³ A sole alkaline phosphatase (ALP) elevation is rarely an indication for further work-up. Personally, I comfortably watch an increased ALP until it exceeds ∼1200 mg/dL and my next step is to complete a minimum database and thyroid panel.

Dogs confirmed to have Cushing’s Syndrome have a constellation of lab work changes that are characterized by liver enzyme elevations, thrombocytosis, a stress leukogram (neutrophilia, lymphopenia, eosinopenia), and highly hyposthenuric urine (<1.008). It is rare for the urine to be concentrated >1.020 in a dog with CS, although it may be isosthenuric (1.008–1.012) if they also have CKD.

The typical “pattern” of liver enzyme elevations includes an ALP elevation with elevated GGT, cholesterol, and triglycerides. The patient should not be icteric and T. bilirubin should be normal. ALT may be normal or mildly elevated (far less so than the ALP). Additional albeit less specific findings might include elevated phosphorous, slight hyperglycemia, low BUN from medullary washout secondary to PU/PD, and spurious hyperkalemia secondary to thrombocytosis.

Both hypertension and clinical proteinuria may be seen in dogs with CS and cases with lab work findings noted above should have blood pressure and urine protein: creatinine performed (assuming clean urine sediment). Dogs with CS may be predisposed to bacterial cystitis and a urine culture might be considered as part of the initial workup. Routine urine culture is not recommended in dogs with stable treated CS unless there are clinical signs of lower urinary tract disease or a significant change in health.⁶

If based on signalment, history, physical examination, and lab work, the clinical suspicion of Cushing’s syndrome is low, further testing is not recommended.

If a screening test must be performed, for instance, because the owner insists, then the test of choice is urine cortisol:creatinine ratio (UCCR) performed on a urine sample collected from home. This test has a high diagnostic sensitivity to rule the disease OUT although it does not confirm the diagnosis and high UCCRs are common in dogs with non-adrenal illness or in stressful environments.

If based on signalment, history, PE, and lab work findings the clinical suspicion of Cushing’s Syndrome is high, a low-dose dexamethasone suppression test is recommended as the diagnostic test.²

Diagnostic Tests

Serum Cortisol Assays

Tests of serum cortisol in veterinary medicine are not nearly as reliable as they seem. First, cortisol levels naturally and sporadically fluctuate in the dog—which complicates determining even a normal reference range. Cortisol results vary between different assays and cannot be directly compared, and cortisol results vary within assays resulting in a degree of imprecision.

For instance, a study of the rapid ELISA snap test showed that intra-test variability was significant enough to impact treatment decisions in 1 of 4 cases rendering the only current widely available in-house cortisol test results questionable, most especially when they are borderline.⁷

Samples for cortisol measurement must be non-hemolytic, non-lipemic, and from serum that was spun off the blood and refrigerated or frozen within 1 hour.³ The patient does not technically need to be fasted for cortisol measurements, however, the risk of obtaining lipemic serum typically dictates a need for the patient to be fasted.

Finally, tests that measure cortisol also measure nearly all synthetic glucocorticoids (*except dexamethasone) which require certain treatments to be discontinued before the test can be performed.

Helpful Tips

• Patients receiving short-term glucocorticoids (<1–2 weeks) must have them discontinued for at least 72 hours before testing.
• Patients receiving chronic (>2 weeks) glucocorticoids, including dexamethasone, must have them tapered and discontinued for at least 3–4 weeks before cortisol testing.
• Patients in need of treatment with glucocorticoids prior to cortisol testing should be treated with dexamethasone and testing should be performed ASAP*

*This is an uncommon scenario in dogs with CS because dogs should not be sick at the time of diagnosis.

Low Dose Dexamethasone Suppression Test (LDDST)

The reported sensitivities of the LDDST in the literature vary from very good to excellent (85–100%) which is far higher than the sensitivity of the ACTH stimulation test (57–83%). This difference is mainly because the ACTH stimulation test performs poorly in dogs with an adrenal tumor. This makes sense when you consider how mutated tumor cells may not respond to ACTH due to altered receptors or cellular signaling processes.

A LDDST is performed by measuring basal cortisol and then administering 0.01 mg/kg dexamethasone IV and measuring cortisol at 4 and 8 hours. A normal dog will respond to exogenous steroids by suppressing natural cortisol production and a normal cortisol 8 hours after dexamethasone should be low. Measuring the cortisol at 4 hours does not impact the diagnosis, however, it is still recommended as it may differentiate ACTH-dependent Pituitary Disease (PDH) from an Adrenal Tumor (AT). Suppression >50% of baseline is consistent with PDH while lack of suppression at 4 hours does not offer any additional useful information.

Accurate dosing of microdoses of IV dexamethasone requires dilution which is best done pre-syringe by creating a 0.1 mg/mL dexamethasone solution (0.5 mL dexamethasone 2 mg/mL + 9.5 mL sterile saline).³ I simply avoid using dexamethasone SP, which is a 3 mg/mL dexamethasone solution (3 mg/mL dexamethasone + 1 mg/mL sodium phosphate) and requires dilution that is too precise to be practical.

ACTH Stimulation Test

The ACTH stimulation test has the highest diagnostic specificity (i.e., ability to rule the disease in) only when the clinical suspicion is truly high. The specificity is not great in dogs with adrenal tumors or in dogs with chronic illness in which high cortisol levels are common and create false positives.

The ACTH stimulation test is typically performed after the LDDST in cases with high clinical suspicion and normal LDDST results.

The ACTH stimulation test is the diagnostic test of choice in dogs with diabetes because the LDDST is more likely to have false negatives in this scenario, meaning we are more likely to miss the diagnosis. However, because the ACTH stimulation test already has a high false-positive rate the clinical suspicion must be high prior to testing otherwise we may falsely diagnose the patient with CS as the cause of their poorly regulated DM.

Indications of possible CS resulting in poorly controlled diabetes include:

• Insulin dose >1 unit/kg
• Persistent PU/PD/PP despite well-controlled glucose
• Persistent PP despite controlled PU/PD
• PE findings consistent with CS but not diabetes (such as comedones, thinning skin, and potbellied appearance)

If the results of the ACTH stimulation test mimic Addison’s disease, meaning the cortisol is <2 at baseline and after ACTH administration, the findings are consistent with iatrogenic Cushing’s. Additional investigation into the source of glucocorticoids is recommended. I have personally documented Iatrogenic CS in dogs being treated with chronic topical ophthalmic dexamethasone, oral budesonide, and inhaled fluticasone.

Differentiating Tests

Differentiating between PDH and AT is important when the owner will pursue definitive therapy, namely surgery. For an adrenal tumor, the most definitive therapy is adrenalectomy. For PDH, hypophysectomy offers the only chance of a cure although abdominal ultrasound is a reasonable part of the workup for Cushing’s disease regardless, as these are typically geriatric patients with multiple comorbidities at the time of diagnosis.

In skilled hands, the adrenals can be reliably evaluated and used to help differentiate PDH from AT.

In PDH the adrenals are typically symmetrically enlarged, however, dogs can develop multiple adrenal tumors, many of which are incidental, and therefore asymmetric adrenal enlargement is also seen. Some dogs with PDH may have “normal” adrenals on ultrasound because there is a range of acceptable measurements published and a large range in dog size.

In cases of an adrenal tumor, assessment of the contralateral adrenal gland is very important, and it should be small and atrophied. Some rare cases have PDH and an adrenal tumor thus complicating the interpretation and indicating a potential need for even further workup.

In my experience in teleradiology, the most common pitfalls in abdominal ultrasound related to the evaluation of CS in dogs include studies that do not identify one or both adrenal glands, studies that have structures measured that are not the adrenal gland, or studies lacking sufficient detail to confirm the adrenal gland.

Adrenal Ultrasound Tips

• Both poles of the adrenals do not have to be seen in one image. It may be easier to measure the poles separately.
• The cranial pole of the right adrenal gland may be especially difficult to see in deep-chested dogs – particularly when it is normal or small.
• Dogs with PDH treated with trilostane will have adrenal glands that continue to grow
• Repeat ultrasound of the adrenals is not indicated in routine monitoring of CS though might be indicated if there is a significant change in the patient’s status.

Treatment

CS is nearly always caused by a tumor and, as with many other types of cancer, the most definitive treatment is surgery. For the practical and financially able client, referral for surgery may provide the best long-term value (i.e., cost relative to outcome and client experience).

Nonetheless, I have found that most of my clients forgo the path ending in surgery. In these cases, a question often posed is “should we treat at all”?

Yes. The median survival time is 500 days even when untreated,⁴ untreated CS results in significant progressive morbidity, dogs who are being treated have higher quality of life scores than those not being treated,⁹ and because nearly 80% of those that die from CS are euthanized.

Dogs who are left untreated for CS will develop additional complications. Examples include protein-losing nephropathy, systemic hypertension, pulmonary thromboembolism (PTE) and pulmonary hypertension, secondary infections (dental, urinary, respiratory, skin and anal sacs, uveitis), cruciate ligament injuries, uroliths, gall bladder mucocele, and diabetes. Many of these complications are prevented or resolve with treatment although proteinuria and hypertension may persist requiring specific concurrent therapy. There is no direct treatment for calcinosis cutis which resolves over time (months) via the normal process of skin overturning and regeneration. Secondary pyoderma and pruritus and/or non-pruritic Demodex may need to be managed with oral and/or topical antibiotics and/or antiparasiticides.

The complications of untreated or poorly controlled CS occasionally result in critical illness such as massive PTE, bile peritonitis secondary to mucocele rupture, sepsis, and diabetic ketoacidosis. In these situations, diagnostic and treatment efforts must be directed at the specific disease process causing illness and not at CS. Trilostane must never be given to the sick pet and attempts to diagnose CS should wait until the patient has fully recovered from illness, at least 2–3 weeks later.

My treatment of choice is name brand Vetoryl®¹⁰ even though I freely admit that I have almost no experience with mitotane which may still be a better choice for some adrenal tumors. Trilostane, the active drug within Vetoryl® is an enzyme inhibitor that blocks the pathway responsible for producing cortisol (and other glucocorticoids). Because this effect relies on adrenocortical cells to respond to the drug, some adrenal tumors may respond poorly to enzyme inhibition.

Studies have shown that compounded trilostane lack consistency and reliability, enough so that in my opinion the potential cost-saving benefits do not make up for the lack of or unreliable control that may result in general or from batch to batch.¹¹ Only trilostane suspended in cod-liver oil has been studied with all other varieties of oral suspensions having unknown absorption, stability, and efficacy characteristics.¹²

The literature overwhelmingly points towards Vetoryl® being a twice or even three times daily drug. Where I once might have looked at once-daily therapy for cost savings, I now recommend putting the money into treatment and being cost conscientious elsewhere (see the section on monitoring below).

I typically start a dose of 1–2 mg/kg every 12 hours which is generally rounded up or down based on capsule size. Consider the lean body mass or ideal body weight of the patient in your calculations, or use the low end of the dose range.

Due to cost, I do my best to avoid doses that require multiple capsule sizes twice daily. This may mean a different dose in the morning than in the evening—for instance 50 mg PO once daily in the morning and 30 mg PO once daily in the evening versus 40 mg (one 30 mg capsule and one 10 mg capsule) twice daily has a potential cost savings of over $100/month.

Monitoring

Perhaps the most used monitoring protocol is published by Dechra, the makers of Vetoryl®¹⁰ although there are many variations on “target” cortisol levels during treatment.

In my opinion, the most important emerging topic in canine CS is that routine ACTH stimulation monitoring is falling out of favor.

First, research has failed to show that the ACTH stimulation test can predict an iatrogenic Addisonian crisis. The incidence of iatrogenic Addison’s disease has proven to be low (15%) and 75% of cases were temporary. The risk did not appear to be related to the Vetoryl® dose and occurred anywhere from 5 days to 4 years after initiating therapy meaning that timed monitoring probably doesn’t make sense.¹³

Further, the cost of synthetic ACTH cannot be ignored. In Europe, synthetic ACTH has become so limited and cost-prohibitive that the European labeling for Vetoryl® has been changed to use the pre-pill cortisol for monitoring; a single cortisol measurement obtained just before the pet’s morning dose is given.¹⁴

Unfortunately, this test, like all tests of cortisol, is unreliable in sick and stressed dogs. Thus, a pre-pill cortisol is most helpful to confirm that the patient is doing well, something that we should already know based on the history and physical examination.¹⁵

As a result, consensus for the best monitoring protocol is still lacking amongst experts, although it is generally accepted that the goal should be a healthy dog and a happy owner.

Presently, I perform an ACTH stimulation test about 30 days after starting a patient on Vetoryl®. This is probably unnecessary although I like to document a baseline (especially if diagnosed with a LDDST), potentially note an initial effect (ACTH stim improved from time of diagnosis), and potentially screen for drug sensitivity or a misdiagnosis (such as a dramatic and rapid response to typical doses).

I typically continue the starting dose after the first recheck and check with the owner regarding clinical signs at 60 days. At that time, if clinical signs persist, a dose increase is indicated and a recheck is scheduled in 30 days. For me, dose increases are typically directed by the capsule sizes. The overall management goal should be to achieve a physical examination that is not consistent with CS meaning that the pet's pot-bellied appearance, alopecia, and hepatomegaly have resolved. The Cushing’s Health-Related Quality of Life Tool can help provide a more objective framework for monitoring.⁹

It isn’t wrong to perform a post-pill ACTH stimulation test if the dog is doing poorly or you are concerned about Addison’s disease even though it has been shown that it neither predicts adrenal crisis nor correlates to clinical improvement. However, little clinically useful information is gained when an ACTH stimulation test is performed in a stable treated dog with CS.

Fortunately, studies have shown that the cost-saving lower dose of 1 ucg/kg of ACTH is appropriate for monitoring (not diagnosis) of dogs being treated for CS.¹⁶

Consider that when the situation dictates a need for monitoring to produce objective data (client concern, practitioner worry, insurance claims), that an immediate pre-pill cortisol of <5 ug/dL can be used to help document good control.¹⁴

If cortisol measurements indicate possible Addison’s disease (<1 mg/dL) in a clinically healthy patient a dose decrease is not necessarily required. The finding might simply be ignored, and the owner reminded about monitoring for clinical signs of Addison’s disease. Otherwise, an ACTH stimulation test performed 9–12 hours after the morning pill can be considered. By exceeding the duration of action of Vetoryl®, measurements performed later in the day often confirm that adrenal reserve is adequate and support that the dog is not iatrogenically Addisonian.¹⁷

References

1.  European Society of Veterinary Endocrinology Project Alive. https://www.esve.org/alive/

2.  Behrend EN, Kooistra KS, Nelson R, et al. Diagnosis of spontaneous canine hyperadrenocorticism: 2012 ACVIM consensus statement (small animal). J Vet Intern Med. 2013;27:1292–1304.

3.  Cook AK. The diagnosis of canine hyperadrenocorticism. Today's Vet Prac. 2019.

4.  Hoffman, JM, Lourenco BN, Promislow DEL, et al. Canine hyperadrenocorticism associations with signalment, selected comorbidities and mortality within North American veterinary teaching hospitals. J Small Anim Prac. 2018;59(11):681–690.

5.  Miceli DD, Pignataro OP, Castillo VA. Concurrent hyperadrenocorticism and diabetes mellitus in dogs. Res Vet Sci. 2017;115:425–431. doi: 10.1016/j.rvsc.2017.07.026.

6.  Weese JS, Blondeau J, Boothe D, et al. International Society for Companion Animal Infectious Diseases (ISCAID) guidelines for the diagnosis and management of bacterial urinary tract infections in dogs and cats. Vet J. 2019;247:8–25.

7.  Lane MB, Flatland B, Olin SJ, et al. Analytic performance evaluation of a veterinary-specific ELISA for measurement of serum cortisol concentrations of dogs. JAVMA. 2018;253(12):1580–1588.

8.  Gould SM, Baines EA, Mannion PA, et al. Use of endogenous ACTH concentration and adrenal ultrasonography to distinguish the cause of canine hyperadrenocorticism. J Small Anim Pract. 2001;42(3):113–121.

9.  Schofield I, O’Neill DG, Brodbelt DC, et al. Development and evaluation of a health-related quality-of-life tool for dogs with Cushing’s syndrome. J Vet Intern Med. 2019;33(6):2595–2604.

10.  www.dechra.com

11.  Cook AK, Bond KG. Evaluation of the use of baseline cortisol concentration as a monitoring tool for dogs receiving trilostane as a treatment for hyperadrenocorticism. J Am Vet Med Assoc. 2010;237(7):801–805.

12.  Crosby J, Brown S. Stability of compounded trilostane suspension in cod liver oil. Vet J. 2017;228:15–17.

13.  King JB, Morton JM. Incidence and risk factors for hypoadrenocorticism in dogs treated with trilostane. Vet J. 2017;230:24–29.

14.   Macfarlane L, Parkin T, Ramsey I. Pre-trilostane and three-hour post-trilostane cortisol to monitor trilostane therapy in dogs. Vet Rec. 2016;179(23):597.

15.  Boretti F, Musella C, Burkhardt W, et al. Comparison of two prepill cortisol concentrations in dogs with hypercortisolism treated with trilostane. BMC Vet Res. 2018;14(1):417.

16.  Aldridge C, Behrend EN, Kemppainen RJ, et al. Comparison of 2 doses for ACTH stimulation testing in dogs suspected of or treated for hyperadrenocorticism. J Vet Intern Med. 2016;30(5):1637–1641.

17.  Midence JN, Drobatz KJ, Hess RS. Cortisol Concentrations in well-regulated dogs with hyperadrenocorticism treated with trilostane. J Vet Intern Med. 2015;29(6):1529–1533.