Approach

The most important decision in the diagnosis of Cushing syndrome is deciding which patients need to begin a diagnostic evaluation. With the growing epidemic of obesity and metabolic syndrome (central obesity with hypertension and insulin resistance), many patients have a Cushingoid phenotype, but most do not have Cushing syndrome.

Important considerations include female sex, as women have a higher proportion of Cushing syndrome, unexplained hypertension (particularly in young patients), new onset of glucose intolerance or diabetes, unexplained weight gain, unexplained fractures (due to premature osteoporosis), hirsutism, menstrual irregularities, and unexplained proximal muscle weakness. Cushing syndrome creates a hypercoagulable state and is associated with an increased risk of venous thromboembolic disease, such as deep vein thrombosis and pulmonary embolism.[1]

History

All patients with suspected Cushing syndrome should have a complete history to exclude the use of oral, injectable, inhaled, or topical glucocorticoids manifesting as iatrogenic disease. At a minimum, patients with the following conditions should go on to be screened:[23][24]

  • Features unusual for age (i.e., osteoporosis or hypertension in young patients)

  • Less-usual features, such as unexplained psychiatric symptoms (including depression)

  • Unexplained nephrolithiasis

  • Multiple or progressive symptoms

  • Polycystic ovary syndrome

  • Pituitary adenomas

  • Adrenal adenomas.[25]

Physical

Clinical features suggesting Cushing syndrome include:

  • Progressive proximal muscle weakness

  • Bruising without obvious trauma

  • Facial plethora or rounding

  • Violaceous striae

  • Supraclavicular fat pad

  • Dorsocervical fat pad.

Children exhibiting weight gain with linear decreased growth velocity should also be considered. In addition, many patients have increased frequency of acne on the face, back, and chest.

Rapid virilisation in females (rapid-onset or increased hirsutism, voice deepening, and clitoral enlargement) in the setting of cortisol excess suggest adrenal carcinoma, which is associated with a 50% to 60% chance of Cushing syndrome.[10]

Initial biochemical testing

One of the following three high-sensitivity tests, or a combination, should be used as a first-line diagnostic test in patients with suspected Cushing syndrome:[23][26][27]

  • Late-night salivary cortisol

  • Overnight 1 mg dexamethasone suppression testing

  • 24-hour urinary free cortisol

The 48-hour 2 mg dexamethasone suppression test is rarely used in isolation but can be used in combination with other tests. Although all of the included diagnostic tests are highly sensitive and specific, dexamethasone suppression testing was found to be the most sensitive while 24-hour urinary free cortisol was less sensitive.[28]

Confirmation of biochemical testing

To increase diagnostic accuracy, the initial high-sensitivity test should be repeated.[23] For example, late-night salivary cortisol samples should be obtained on two separate nights.[29][30][31] Similarly, at least two 24-hour urinary free cortisol samples should be collected.[32] Alternatively, a second high-sensitivity test may be performed: for example, supplementing a late-night salivary cortisol test with a 24-hour urinary free cortisol measurement.[33] Testing of late-night salivary cortisol is more accurate at initial diagnosis, but late-night salivary cortisol can frequently be normal in patients with recurrent/persistent disease after pituitary surgery, requiring more samples for testing.[34]

Patients with initial normal biochemical testing are unlikely to have Cushing syndrome. If signs or symptoms progress, or intermittent Cushing syndrome is suspected, repeat biochemical testing can be performed after 6 months or at a time when cortisol hypersecretion is assumed.[23]

Patients with any abnormal initial biochemical testing require further investigation and referral to an endocrinologist should be considered. Before proceeding with any further evaluation, physiological causes of hypercortisolism should be excluded. These conditions include: psychiatric disorders including depression, alcohol use disorder, physical stress, malnutrition, pregnancy, and perhaps class III obesity (BMI 40 or above) or metabolic syndrome.[26][35][36][37] Urine pregnancy testing should be considered to exclude pregnancy, and glucose testing may reveal concomitant glucose intolerance or diabetes.

If multiple additional tests are abnormal, the patient has Cushing syndrome and differential diagnostic testing should be undertaken. Greatly elevated dehydroepiandrosterone sulfate levels are suggestive of adrenocortical carcinoma, but are neither sensitive nor specific.

Algorithm for diagnosis of Cushing syndrome Opens in new window[26]

Initial differential diagnostic testing

Once hypercortisolism has been established, further testing to determine the aetiology should be sought.

Morning plasma adrenocorticotrophic hormone (ACTH) is the test of choice for differentiating ACTH-dependent from ACTH-independent Cushing syndrome. Suppressed ACTH levels (<1 picomol/L [<5 picograms/mL]) suggest ACTH-independent Cushing syndrome (although assays differ between different laboratories), and further investigations should include imaging of the adrenal glands to identify adrenal pathology causing hypercortisolism, such as an adenoma.

Unsuppressed ACTH levels (>4 picomol/L [>20 picograms/mL]) suggest ACTH-dependent Cushing syndrome (although assays differ between different laboratories).[1] Further investigations in such patients should include pituitary/sellar magnetic resonance imaging (MRI) to identify an ACTH-secreting pituitary adenoma. A spoiled-gradient echo 3D T1 sequence has been shown to have higher sensitivity than dynamic MRI for detecting and localising pituitary micro-adenomas in patients with Cushing syndrome.[38]

Patients with ACTH-dependent Cushing syndrome and an adenoma ≥10 mm on MRI should proceed to treatment. Some clinicians prefer additional biochemical confirmation with high-dose dexamethasone suppression testing before initiating surgical therapy.[39] The use of high-dose dexamethasone suppression testing is an area of debate because of its variable sensitivity and specificity.[40]

Inferior petrosal sinus sampling

Patients with ACTH-dependent Cushing syndrome have either Cushing's disease (90% to 95%) or ectopic ACTH production (5% to 10%). Those without definitive lesions on MRI should undergo inferior petrosal sinus sampling (IPSS); this should be carried out in a specialised centre because of potential patient risk.[26] Patients with adenomas 6-9 mm on MRI should have IPSS to confirm the diagnosis.[26] Up to 40% of patients with Cushing's disease will not have visible lesions on pituitary/sellar MRI. Patients without definitive lesions on MRI should also undergo inferior petrosal sinus sampling.[1][41] IPSS is the only test with sufficient diagnostic accuracy to differentiate Cushing's disease from ectopic ACTH production.[42][43] Patients with an IPSS central/peripheral gradient >2:1 or 3:1 after corticotrophin-releasing hormone (CRH) stimulation have Cushing's disease and can undergo therapy. Those without a gradient are likely to have ectopic ACTH production.[41]

Tests for ectopic ACTH syndrome

Patients without an IPSS central/peripheral gradient >2:1 or 3:1 after CRH stimulation should be investigated for ectopic ACTH secretion. This evaluation generally includes computed tomography (CT) scanning of the chest, abdomen, and pelvis to look for a tumour secreting ACTH. The most common tumours that secrete ACTH are bronchial or thymic carcinoids. Other neuroendocrine tumours include islet cell and medullary thyroid cancer.[43] An MRI of the chest may be helpful in selected cases; other imaging modalities such as fluorodeoxyglucose positron emission tomography, gallium-68 DOTATATE positron emission tomography/CT, and octreotide scanning may be considered in some patients.[44][45][46][47][48]

Mild autonomous Cushing syndrome

These patients have an incidentally discovered adrenal adenoma producing mildly excessive cortisol (previously known as subclinical Cushing syndrome). Mild cortisol excess requires a specialised approach and may be more difficult to diagnose as a function of the inherent nature of the disease.[49] In patients with incidentally discovered adrenal nodules without clinical features of Cushing syndrome, use the 1 mg dexamethasone suppression test as the initial diagnostic test because urinary free cortisol and late-night salivary cortisol have a lower sensitivity in these patients.​[27][50][51]​​[52]

One meta-analysis has shown that patients with bilateral adrenal incidentalomas present a higher prevalence of subclinical Cushing syndrome compared with patients with unilateral adrenal incidentalomas.[53]

Use of this content is subject to our disclaimer