I. Definition.

Cushing’s syndrome is a clinical condition resulting from prolonged exposure to excessive glucocorticoids, either from endogenous or exogenous sources. Therapeutic administration of exogenous glucocorticoids is the most common cause of Cushing’s syndrome. The most common endogenous cause of Cushing’s syndrome is Cushing’s disease.

Cushing’s disease is Cushing’s syndrome caused by an adrenocorticotropic hormone (ACTH) secreting pituitary adenoma. Other endogenous causes of Cushing’s syndrome include ectopic secretion of ACTH or corticotropin-releasing hormone (CRH) by nonpituitary tumors and primary adrenal tumors.

II. Diagnostic Approach

A. What is the differential diagnosis for this problem?

  • Most common cause is iatrogenic Cushing’s syndrome caused by exogenous glucocorticoids

  • Cushing’s disease (60-70%): pituitary adenoma (usually microadenoma) or hyperplasia

  • Adrenal tumor (15-25%): adenoma or carcinoma

  • Ectopic ACTH (5-10%): SCLC, carcinoid, islet cell tumors, medullary thyroid cancer, pheochromocytoma

Cushing’s syndrome can broadly be classified into either ACTH dependent or ACTH independent etiology. A third classification of pseudo-Cushing’s syndrome is also considered in the appropriate clinical setting. SeeTable I for Causes of Cushing’s syndrome according to underlying pathology.

Table I.
Cushing’s disease (pituitary hypersecretion of ACTH) 70
Ectopic ACTH syndrome (non pituitary tumors) 10
Ectopic CRH syndrome (nonhypothalamic tumors causing pituitaryhypersecretion of ACTH) <1
Adrenal Adenoma 10
Adrenal carcinoma 5
Micronodular hyperplasia 1
Macronodular hyperplasia <2
Major- depressive disorder 1
Alcoholism <1

B. Describe a diagnostic approach/method to the patient with this problem

Establishing the diagnosis of Cushing’s syndrome involves three steps:

  • Suspecting Cushing’s syndrome on the basis of the patient’s symptoms and signs.

  • Documenting the presence of hypercortisolemia.

  • Determining the cause of hypercortisolism.

1. Historical information and Physical Examination important in the diagnosis of Cushing's syndrome.

Patients with Cushing’s syndrome may present with a wide spectrum of manifestations ranging from subclinical to overt syndrome, depending on the duration and intensity of excess glucocorticoid production, the presence or absence of mineralocorticoid and androgen excess, and the cause of hypercortisolism.

Symptoms of Cushing’s syndrome that occur in the setting of malignancy may be overshadowed by symptoms of malignancy e.g. weight loss instead of weight gain. Patients with cyclical Cushing’s syndrome may have intermittent symptoms. Many of the clinical manifestations of hypercortisolemia tend to be less severe in older patients (>50 years). Some of the incidentally discovered adrenal adenomas have subclinical Cushing’s syndrome associated with glucose intolerance and hypertension.

It is important to note that none of the major findings suggestive of Cushing’s syndrome are truly pathognomonic of the syndrome, and therefore a high degree of clinical suspicion is required to diagnose the condition.

Progressive central obesity

Involving the abdomen, face and neck (buffalo hump, moon facies, supraclavicular fat pad and exophthalmos from retroorbital fat deposition).

Metabolic complications

Glucose intolerance (gluconeogenesis stimulation by cortisol and peripheral insulin resistance due to obesity), hypertension (through poorly understood multifactorial etiologies), both these increase cardiovascular risk, a major cause of morbidity and mortality in patients with Cushing’s syndrome. Severe hypertension and hypokalemia commonly seen in patients with ectopic ACTH secretion (excess cortisol overwhelms the capacity of the 11-b-hydroxy steroid dehydrogenase type II, which oxidizes cortisol to inactive cortisone, thereby resulting in activation of mineralocorticoid receptors).

Dermatological manifestations

Skin atrophy, fragile skin with easy bruisability, wide purple striae and cutaneous fungal infections. Edema and hyperpigmentation may be seen with ectopic ACTH syndrome.

Reproductive changes

Include menstrual irregularities, hirsutism, oily facial skin with acne, signs of virilization e.g. temporal balding, deepening of voice, especially in women with adrenal carcinoma.

Musculoskeletal manifestations

Includes proximal myopathy, muscle wasting (resulting from catabolic effect of excess glucocorticoid on skeletal muscles), osteoporosis (caused by decreased bone formation, increased bone resorption, decreased intestinal and renal calcium absorption), vertebral compression fractures, pathological fractures of the ribs and long bones, aseptic necrosis of the femoral heads that may be present.

Neuropsychiatric changes

Include labile mood, agitated depression, anxiety, panic attacks, mild paranoia, impaired short-term memory and cognition, and insomnia.

SeeTable II for signs and symptoms in patients with Cushing’s syndrome.

Table II.
Sign/symptom %Patients
Truncal Obesity 95%
Facial Fullness 90%
Gonadal dysfunction 90%
Skin Atrophy and bruising 80%
Menstrual irregularity 80%
Hypertension 75%
Hirsutism, acne 75%
Mood disorder 70%
Muscle weakness 65%
Diabetes or glucose intolerance 60%
Osteopenia, osteoporosis, fractures 50%

3. Laboratory, radiographic and other tests that are likely to be useful in diagnosing the cause of this problem.

Biochemical diagnosis of Cushing’s syndrome needs to be established before initiating imaging studies. The biochemical diagnosis of Cushing’s syndrome involves three critical steps:

  • Documenting the presence of hypercortisolism i.e. does the patient have Cushing’s syndrome?

  • Determining if the cortisol excess is ACTH independent or ACTH dependent i.e. does the patient have primary adrenal disease or an ACTH-secreting tumor?

  • Determining the source of the ACTH in the ACTH dependent etiology of Cushing’s syndrome i.e. does the patient have Cushing’s disease or ectopic ACTH syndrome?

SeeFigure 1

Figure 1.

A simplified paradigm for screening patients for Cushing’s syndrome using late-night salivary cortisol.

C. Criteria for Diagnosing Each Diagnosis in the Method Above.

Does the patient have hypercortisolism i.e. does the patient have Cushing’s syndrome?

Before evaluation for possible Cushing’s syndrome a careful history should be acquired, excluding exogenous causes of glucocorticoid intake and including glucocorticoids taken orally, injected glucocorticoids, topical steroid applications, and inhaled glucocorticoids.

At least two first-line tests should be abnormal to establish the diagnosis of Cushing’s syndrome. Late night salivary cortisol, urinary cortisol, and the low-dose dexamethasone suppression tests are recommended as first-line tests. The diagnosis of Cushing’s syndrome is confirmed when two tests are unequivocally abnormal.

a) 24-hour urine free cortisol measurement is a very good screening test for Cushing’s syndrome as it provides the most direct and reliable practical index of cortisol secretion, given that it is an integrated measure of the serum-free cortisol concentration. A complete 24-hour specimen and a reliable reference laboratory are critical for obtaining a valid result.

The patient can be assumed to have Cushing’s syndrome if on at least two separate occasions basal urinary cortisol excretion is two to three fold higher than the normal range. If the 24-hour urinary cortisol is equivocally increased the patient needs to be reevaluated in several weeks or undergo further testing to differentiate between mild Cushing’s, cyclical Cushing’s and pseudo Cushing’s syndrome.

b) The low-dose dexamethasone suppression tests (LDDST) is another standard screening test commonly used test to differenciate between patients with Cushing’s syndrome from any cause from patients who do not have Cushing’s syndrome. Exogenous dexamethasone dosages used should reliably suppress ACTH secretion by the normal pituitary gland, leading to suppression of cortisol secretion, subsequent reductions in serum cortisol concentrations and urinary excretion of cortisol and cortisol metabolites.

There are two forms of the low-dose dexamethasone suppression test: the 1mg “overnight” and the two-day 2mg test:

  • Overnight 1mg test — The overnight test consists of administration of 1.0mg of dexamethasone at 11PM to 12AM, and measurement of serum cortisol at 8AM the next morning. Using current specific immunoassays, most normal individuals have an 8AM serum cortisol value of less than 1.8mcg/dL.

  • Standard two-day 2mg test — The two-day 2mg test consists of administering 0.5mg of dexamethasone every six hours for eight doses, and measurement of serum cortisol either two or six hours after the last dose. Most normal individuals have a cortisol level suppression of less than 1.8mcg/dL.

Since LDDST has a high false positive rate, it should not be used as the sole criterion for making the diagnosis of Cushing’s syndrome. In case of a positive test, it may be informative to retain a serum sample collected at the time of cortisol measurement for measurement of the dexamethasone level. A low dexamethasone level suggests non-compliance, individual variation in dexamethasone metabolism, or individual drug variability effects on dexamethasone metabolism. The dexamethasone suppression tests are not a good choice for patients taking medications that may alter the metabolism of the drug.

c) Late evening salivary cortisol – Measurement of serum or salivary cortisol in the late evening is based upon the fact that the normal evening nadir in serum cortisol is preserved in obese and depressed patients, but not in those with Cushing’s syndrome. Late evening salivary cortisol measurement has distinct advantages: it is noninvasive, saliva is easily collected, cortisol is stable in saliva at room temperature for several days, and can be performed by patient at home.

It is especially helpful in patients suspected of having cyclical or intermittent Cushing’s syndrome. Late evening salivary cortisol measurement is a usefull test for the diagnosis of Cushing’s syndrome, however appropriate assay-specific and perhaps age-specific normative values must be used for its interpretation.

d) Late evening serum cortisol – This test has usually been performed in the hospital. The test can reliably beperformed on an ambulatory basis, aswell as asking the patient to return between 11PM and midnight for blood-drawing. At least two samples are required for assessment. Patients with intermediate values should be reevaluated after several weeks.

e) CRH after dexamethasone test – This is helpful in distinguishing patients with Cushing’s syndrome from those with peudo-Cushing’s syndrome. Compared to patients with Cushing’s syndrome, depressed patients continue to show a suppressed plasma cortisol level even after infusion of CRH, which reflects preserved sensitivity of ACTH secretion to dexamethasone suppression.

SeeTable III for tests useful in diagnosing Cushing’s syndrome.

Table III.
1mg overnight DST Dex, 1mg PO at 11PM 8AM plasma cortisol Normal, <5mcg/dl 98 70-80
24-hour UFC 24-hour urine collection Cortisol, creatinine Values 2- to 3-fold higher than the upper limit of normal suggests Cushing’s syndrome 95-100 98
2-day LDDST (urine) Dex, 0.5mg PO q6hr for 48 hours (last dose 6AM) 24 hour urine for cortisol, 17-OHCS, creatinine during last 24 hour of dex administration UFC >36mcg/day or urine 17-OHCS >4 mg/day suggests Cushing’ssyndrome 56-69 74-100
2-day LDDST (serum) Dex, 0.5mg PO q6hr for 48 hours (last dose 6AM) Plasma cortisol 2 hours after last dose of dex Normal <1.4mcg/dl 90 100
CRH/dex Same as LDDST with first dose dex given at noon; CRH, 1mcg/kg IV at 8AM after last dose of dex Plasma cortisol 15 mins after CRH injection Cortisol >1.4mcg/dl suggests Cushing’s syndrome 100 100
Late evening salivary cortisol 11PM sample Salivary cortisol Value <1.3ng/ml excludes Cushing’s syndrome 92 100
Midnight plasma cortisol Indwelling catheter and hospitalization recommended Midnight plasma cortisol Cortisol >7.5mcg/dl suggests Cushing’s syndrome 96 100
Is the Cushing's syndrome ACTH independent or ACTH dependent i.e. does the patient have primary adrenal disease or an ACTH-secreting tumor?

Once the biochemical diagnosis of hypercortisolism is established, the second stage of the work-up involves the differentiation whether the hypercortisolism is ACTH-dependent (i.e., due to a pituitary or non-pituitary ACTH-secreting tumor), or ACTH independent (i.e., due to anadrenal source) by measuring plasma ACTH. This test is now best performed using a two-site immunoradiometric assay (IRMA).

Plasma ACTH concentrations are normally between 20 and 80pg/mL at 8AM. Between 11PM and midnight or at bedtime the ACTH concentration is normally nearing or at the nadir, usually less than 5pg/mL. Plasma ACTH measurements at any time of the day are satisfactory in patients in whom hypercortisolism has been established because they lack a normal circadian rhythm. Measurement on at least two separate days is recommended.

  • Undetectable or low levels of ACTH (<5mcg/dl) in a patient with serum cortisol concentration greater than 15mcg/dl characterizes a primary adrenal source (ACTH independent Cushing’s syndrome). Thin-section computer tomography (CT) or magnetic resonance imaging (MRI) of the adrenal glands looking for an adrenal mass is usually the next procedure.

  • ACTH levels greater than 15mcg/dl typically indicate an ACTH dependent cause, and the next step involves a search for the source of the high ACTH (pituitary versus non-pituitary i.e. ectopic ACTH or CRH secretion).

  • An ACTH level in the 5 to 15mcg/dl range is considered indeterminate and should be repeated. In most cases the cortisol secretion turns out to be ACTH dependent. If the ACTH level remains consistently in the indeterminate range, a CRH- stimulation test can be helpful. After CRH infusion, if the peak ACTH response is blunted (<10mcg/dl), an ACTH independent (adrenal) cause is likely. However if there is a greater than 50% increase in mean plasma cortisol from baseline after CRH infusion, an ACTH- dependent cause is likely.

What is the source of the ACTH in the ACTH-dependent Cushing's syndrome i.e. does the patient have Cushing's disease orectopic ACTH syndrome?

Making a distinction between patients with Cushing’s disease and ectopic ACTH syndrome is critical as it significantly influences treatment. Several tests can be utilized to differentiate between pituitary Cushing’s disease and non-pituitary (ectopic) ACTH syndrome.

One fundamental difference between the two conditions is the fact that Cushing’s disease is usually suppressible by dexamethasone and stimulatable by CRH, in contrast to ectopic ACTH syndrome. Therefore the commonly used initial test is the high dose dexamethasone suppression test (HDDST), which takes advantage of the fact that ACTH secretion by pituitary adenoma is only relatively resistant to negative feedback regulation by glucocorticoids.

Non-pituitary tumors on the other hand producing ectopic ACTH are completely resistant to such feedback (with the exception of bronchial carcinoids). The high dose test can be administered as a standard 2-day test (2mg, every 6 hours x 8 doses) or a single 8mg dose. If the urine free cortisol suppresses more than 90%, or the plasma cortisol suppresses over 50% from baseline, Cushing’s disease is most likely. An MRI of the pituitary gland should be obtained next.

If Cushing’s disease can not be definitely differenciated from ectopic ACTH syndrome with HDDST and MRI, additional studies may be useful. Inferior petrosal sinus sampling (IPSS) may demonstrate pituitary hypersecretion of ACTH by documenting a central-to-peripheral ACTH gradient in the blood draining the tumor. If there is no conclusive central-to-peripheral gradient after IPSS, the search for an ectopic ACTH source must continue.

Octreotide or pentetreotide scintigraphy can detect some ectopic ACTH secreting tumors, although neither is specific for ACTH-secreting tumors. Chest and upper abdominal CT or MRI should be performed to identify masses or to confirm any positive pentetreotide scans. PET scan is not recommended for routine use.

SeeTable IV for tests to differentiate between Cushing’s disease from ectopic ACTH syndrome.

Table IV.
2-day HDDST Collect 24-hour urine for 17-OHCS or UFC, creatinine; give dex, 2mgPO q6hr for 48 hours; recollect urine during last 24 hours of dexadministration Suppression of urine 17-OHCS >64% or UFC >90% compared to baseline suggests Cushing’s disease. 83 100
8mg overnight dex Measure plasma cortisol at 8am; give dex, 8mg p.o. at 11pm; re-draw blood for plasma cortisol next morning at 8AM. Suppression of plasma cortisol by > 50% compared to baselinesuggests Cushing’s disease. 92 100
CRH Place an in dwelling venous catheter 2 hours prior to testing. Give CRH 1 mcg/kg IV bolus at 8AM; draw plasma ACTH at -5, -1 minute before CRH and +15, +30 minutes after CRH. Increase at ACTH (mean of + 15 and +30 minute values) by 35% greater than baseline (mean of -5 and -1 minute) suggests Cushing’s disease. ACTH secreting tumors do not respond. 93 100
IPSS Simultaneous bilateral inferior petrosal sinus sampling and peripheral sampling for ACTH before and after CRH, 100mcg IV. Basal petrosal: peripheral ACTH ratio >2 or post CRH petrosal: peripheral ACTH ratio >3 suggests Cushing’s disease 97- 100 100

*17-hydroxycorticosteroids (17-OHCS); adrenocorticotropic hormone (ACTH); corticotrophin-releasing hormone (CRH); dexamethasone (dex); high dosedexamethasonesuppression test (HDDST); inferior petrosal sinus sampling (IPPS); urine free cortisol (UFC).

III. Management while the Diagnostic Process is Proceeding

A. Management of Cushing's syndrome.

The treatment is entirely based on the source of the hypercortisolism. Ideal therapy of Cushing’s syndrome would achieve the following goals:

  • Reverse the clinical manifestations by reducing cortisol secretion to normal

  • Eradicate any tumor threatening the health of the patient

  • Avoid permanent dependence upon medications

  • Avoid permanent hormone deficiency

In certain patients, however, one or more of the last three goals may have to be sacrificed to achieve the essential first goal. Specific treatment may be delayed during diagnostic testing or with drug adjustments. During this time, treatment of co-morbidities such as hypertension, osteoporosis and diabetes should be instituted, and the use of medications to prevent thrombosis or bone loss should be considered.

Exogenous Cushing's syndrome

The treatment of Cushing’s syndrome due to exogenous therapy is to stop the glucocorticoid. Gradual withdrawal of glucocorticoids is important, because most patients on long-term therapy with have some degree of hypothalamic-pituitary adrenal – axis suppression with resultant adrenal insufficiency if therapy is abruptly discontinued.

Adrenocorticotropic hormone independent Cushing's syndrome

Adrenal imaging with CT or MRI will demonstrate unilateral or bilateral adrenal disease. Patients with adrenal adenomas or carcinomas are treated with unilateral adrenalectomy. Bilateral adrenalectomy is recommended for patients with bilateral micronodular or macronodular disease. During and after unilateral adrenalectomy, patients should receive glucocorticoid replacement until the hypothalamic-pituitary adrenal axis recovers from the prolonged suppressive effects of glucocorticoid excess. Patients with bilateral adrenalectomy require lifelong glucocorticoid and mineralocorticoid replacement. In addition to adrenalectomy, patients with adrenal carcinomas typically require medications to control hypercortisolemia, which frequently recurs.

Cushing's disease

Transsphenoidal microadenectomy is the treatment of choice for patients with Cushing’s disease with a clearly circumscribed microadenoma. Subtotal pituitary resection can be considered when no adenoma is found during surgery and subtotal (85%-90%) resection of the anterior pituitary gland may be performed if future fertility is not desired for the patient.

Pituitary irradiation is an option for patients in whom fertility is an important concern and in whom a tumor is not found or who are not cured by transsphenoidal resection of a tumor. Patients may undergo pituitary irradiation with either conventional radiation or stereotactic radiation. Pituitary irradiation may not control the hypercortisolemia for months to years and the patients during this time period treated with one or several adrenal enzyme inhibitors medical therapy until the full effects of the radiation are seen.

Adrenalectomy is bilateral total adrenalectomy with lifelong daily glucocorticoid and mineralocorticoid replacement therapy is the final definitive cure, and may be preferred by some patients instead of radiation therapy. Bilateral adrenalectomy may need to be performed in patients with refractory disease.

Nelson’s syndrome occurs in up to 25% of patients with Cushing’s disease who are treated with bilateral adrenalectomy. In this syndrome, the pituitary enlarges and the patient develops hyperpigmentation associated with high ACTH levels. Pituitary irradiation before bilateral adrenalectomy may prevent Nelson’s syndrome.

Surgical cure with transsphenoidal adenomectomy can be assessed with post-operative morning cortisol and ACTH levels, which are undetectable with successful complete resection of the tumor. During and after the transsphenoidal resection of the adenoma, patients require glucocorticoid replacement until recovery of the hypothalamic-pituitary- adrenal axis.Cytotoxic chemotherapy may offer temporary remission in patients with aggressive tumors.

Ectopic adrenocorticotropic hormone or corticotropin-releasing hormone syndrome

Tumors that can be localized by imaging studies should be removed surgically. Imaging studies include CT, MRI and somatostatin receptor scintigraphy. Areas that may be imaged for the presence of tumor include the chest, abdomen, pelvis and neck. If the source is an occult tumor or if there is metastatic disease, medical treatment is required to control the hypercortisolism with adrenal enzyme inhibitors such as ketoconazole, metyrapone and etomidate. Tumor suppressing somatostatin receptors may be responsive to somatostatin analogs. Bilateral adrenalectomy may be performed in refractory cases.

Medical therapy with adrenal enzyme inhibitors

Patients treated with these agents should have their plasma cortisol and 24-hour urine free cortisol levels monitored and doses titrated to keep cortisol levels in the normal range. Glucocorticoid replacement is added as cortisol levels approach the low-normal range.

Ketoconazole is an antifungal agent that inhibits 17-20 lyase, 11-B-hydroxylase and cholesterol side chain cleavage enzyme. It is usually the first-line medication. It’s cortisol-reducing effects are dose dependent and can be seen rapidly. The major side effect is liver toxicity. Other side effects include gynecomastia, impotence and gastrointestinal symptoms. Doses range from 200mg to 1200mg orally daily in two to three divided doses.

Mitotane inhibits cholesterol side chain cleavage enzyme and 11-B-hydroxylase. Mitotane induces permanent destruction of adrenocortical cells and therefore can be used to achieve medical adrenalectomy as an alternative to surgical adrenalectomy. Glucocorticoid replacement is started at initiation of mitotane treatment. Mineralocorticoid treatment may eventually be required. Side effects are eventually dose dependent and include gastrointestinal symptoms, weakness, lethargy, leukopenia, gynecomastia and hypercholesterolemia. Doses start at 0.5 grams orally at bed time and are increased slowly to 2 to 3 gram/day in 3 to 4 divided doses for a total of 6 to 9 months.

Metyrapone inhibits 11-B hydroxylase. Major side effects include increased androgens, hypertension and hypokalemia secondary to increased 11-deoxycorticosterone. Doses range from 250 to 1000mg orally, given every 6 hours. Lower doses of 500 to 750mg orally daily can be used when given in combination with ketoconazole and/or aminoglutethimide.

Aminoglutethimideis an anticonvulsant that inhibits the cholesterol side chain cleavage enzyme. The usual dose is 250mg 2 to 3 times a day. Side effects include gastrointestinal symptoms, lethargy, ataxia, hypothyroidism, head ache, bone marrow supression, and skin rash. It is not as effective as monotherapy compared with the above mentioned agents and therefore is used in combination with other agents.

Medical therapy with other agents

Mifipristone (RU486) is an antiprogestin, a cortisol receptor antagonist at high doses.Octreotide may be useful for ectopic ACTH syndrome.Etomidate is the only intravenous drug that can reduce cortisol levels. May be useful in hospitalized patients for acute management when used in non sedation doses as continuous infusion.Bromocriptine has a modest effect in Cushing’s disease.


Untreated Cushing’s syndrome is often fatal, with most deaths occurring from cardiovascular/thromboembolic or bacterial/fungal infections. Effective therapy, either by surgical cure or pharmacologic control of hypercortisolism, leads to gradual improvement in the clinical manifestation of Cushing’s syndrome over a period of a few months to a year. Hypertension, glucose intolerance, osteoporosis and psychiatric symptoms generally improve, but may not resolve completely.

Cure rates after transsphenoidal surgery have been documented ranging up to 80 to 90%. Remission is achieved in an additional 45 to 80% of patients undergoing pituitary irradiation after unsuccessful pituitary surgery. Patients requiring a second pituitary surgery have a remission rate of 43 to 71%. Most patients become hypoadrenal for months to years after surgery, requiring glucocorticoid replacement surgery.

It is hard to predict the degree of residual pituitary function after partial hypophysectomy. The more extensive the resection, the greater the risk of loss of pituitary function. Long-term follow-up and monitoring for clinical manifestation of tumor recurrence is mandatory. The hypothalamic-pituitary adrenal axis must be evaluated 6 to 12 months after surgery to determine the potential need for exogenous steroid replacement therapy. Patients with panhypopituitarism secondary to surgery require lifelong monitoring and titration of hormone replacement therapy.

Adrenal adenomas and benign ACTH secreting tumors that can be completely surgically resected can be cured. Patients with ectopic ACTH secretion and adrenocortical carcinoma may have a poor prognosis with the underlying malignancy. The patients are rarely cured, but the hypercortisolemia can be controlled with medications or bilateral adrenalectomy. These patients have mortality secondary to the malignancy rather than the hypercortisolemia.

What's the evidence?

Newell-Price, J. “Cushing's syndrome”. Lancet. vol. 367. 2006. pp. 1605-1617.

Chakravarthy, MV. “A simplified paradigm for screening patients for Cushing’s syndrome using late-night salivary cortisol (cutoffs for salivary cortisol from Ref. 5)”. The Journal of Clinical Endocrinology & Metabolism. vol. 91. 2006. pp. 3746-3753.