The manifestations of Cushing syndrome vary considerably. Early in the course of the disease, patients frequently complain of nonspecific symptoms, such as fatigue or reduced endurance but may have few, if any, of the physical stigmata described below. On average, patients with Cushing syndrome have seen four clinicians with a delay in diagnosis of 5 years before the diagnosis is made. Central obesity with a plethoric “moon face,” “buffalo hump,” supraclavicular fat pads, protuberant abdomen, and thin extremities eventually develop in most patients with Cushing syndrome. Muscle atrophy causes weakness, with difficulty standing up from a seated position or climbing stairs. Patients may also experience backache, headache, hypertension, osteoporosis, avascular necrosis of bone, acne, superficial skin infections (eFigure 26–24), and oligomenorrhea or amenorrhea in women or erectile dysfunction in men. Patients may have thirst and polyuria (with or without glycosuria), renal calculi, glaucoma, purple striae (eFigure 26–25) (especially around the thighs, breasts, and abdomen), and easy bruisability. Unusual bacterial or fungal infections are common. Wound healing is impaired. Mental symptoms may range from diminished ability to concentrate to increased lability of mood to frank psychosis. Patients are susceptible to opportunistic infections. Hyperpigmentation is common with ectopic ACTH-secreting neoplasms that tend to produce very high plasma ACTH levels; hyperpigmentation is uncommon with pituitary Cushing disease.
Typical findings in Cushing syndrome. (Reproduced, with permission, from McPhee SJ et al [editors]: Pathophysiology of Disease: An Introduction to Clinical Medicine, 7th ed. McGraw-Hill, 2014.)
Axillary pigmented striae of Cushing syndrome preoperatively. (Used, with permission, from C Grunfeld, MD.)
Adrenal carcinomas usually have gross metastases by the time of diagnosis. Microscopic metastases are not visible by scanning but can be inferred from the presence of detectable cortisol levels following removal of the primary adrenal tumor in patients with a cortisol-secreting carcinoma and Cushing syndrome. The ENSAT staging system is used: stage 1 is a localized tumor 5 cm or smaller; stage 2 is a localized tumor larger than 5 cm; stage 3, tumor with local metastases; and stage 4, tumor with distant metastases.
Glucose tolerance is impaired as a result of insulin resistance. Polyuria is present as a result of increased free water clearance; diabetes mellitus with glycosuria may worsen it. Patients with Cushing syndrome often have leukocytosis with relative granulocytosis and lymphopenia. Hypokalemia may be present, particularly in cases of ectopic ACTH secretion.
1. Diagnostic tests for hypercortisolism
Testing for hypercortisolism involves determining whether the following characteristics of Cushing syndrome are present: (1) lack of cortisol diurnal variation, (2) reduced suppressibility of cortisol by dexamethasone, (3) increased cortisol production rate, and (4) suppression of plasma ACTH by hypercortisolism from an adrenal nodule. Conflicting results are common.
Late-night (10–11 PM) salivary cortisol determinations are particularly useful, especially for ACTH-dependent hypercortisolism. Assays are available that use liquid chromatography-tandem mass spectrometry. Late-night salivary cortisol levels are normally 150 ng/dL (4.0 nmol/L) or less. Late-night salivary cortisol levels that are consistently greater than 250 ng/dL (7.0 nmol/L) are considered very abnormal. The late-night salivary cortisol test has a relatively high sensitivity and specificity for Cushing syndrome. Midnight serum cortisol levels greater than 7.5 mcg/dL (200 nmol/L) are indicative of Cushing syndrome and distinguish it from other conditions associated with a high urine free cortisol (pseudo-Cushing states). Requirements for this test include being in the same time zone for at least 3 days, being without food for at least 3 hours, and having an indwelling intravenous line established in advance for the blood draw.
The overnight dexamethasone suppression test is an easy screening test for hypercortisolism and is particularly sensitive for mild ACTH-independent hypercortisolism from an adrenal nodule. Dexamethasone 1 mg is given orally at 11 PM and serum is collected for cortisol determination at 8 AM the next morning; a cortisol level less than 1.8 mcg/dL (50 nmol/L, high-performance liquid chromatography [HPLC] assay) excludes Cushing syndrome with some certainty. However, 8% of established patients with pituitary Cushing disease have dexamethasone-suppressed cortisol levels less than 2 mcg/dL (55 nmol/L). Antiseizure drugs (eg, phenytoin, phenobarbital, primidone) and rifampin accelerate the metabolism of dexamethasone, causing a lack of cortisol suppression by dexamethasone. Estrogens—during pregnancy or as oral contraceptives or HRT—may also cause lack of dexamethasone suppressibility.
A 24-hour urinary free cortisol and creatinine is usually used to confirm hypercortisolism in patients with a high late-night salivary cortisol or an abnormal dexamethasone suppression test. A high 24-hour urine free cortisol (greater than 50 mcg/day or 140 nmol/day in adults), or free cortisol to creatinine ratio of greater than 95 mcg cortisol/g creatinine, helps confirm hypercortisolism. However, many patients with mild hypercortisolism have a urinary free cortisol that is misleadingly within the reference range when measured by liquid chromatography-tandem mass spectrometry. A misleadingly high urine free cortisol excretion occurs with high fluid intake. In pregnancy, urine free cortisol is increased, while 17-hydroxycorticosteroids remain normal and diurnal variability of serum cortisol is normal. Carbamazepine and fenofibrate cause false elevations of urine free cortisol when determined by HPLC.
2. Diagnostic tests for the source of hypercortisolism
Once hypercortisolism is confirmed, a plasma ACTH and plasma DHEAS are obtained. The plasma ACTH must be collected properly in a plastic tube on ice and processed quickly by a laboratory with a reliable, sensitive assay. A plasma ACTH below 6 pg/mL (1.3 pmol/L), with a low serum DHEAS, indicates a probable adrenal tumor, whereas higher levels are produced by pituitary or ectopic ACTH-secreting tumors. Certain ACTH assays suffer interference and report low-normal plasma ACTH levels in patients with ACTH-independent hypercortisolism. Serum dehydroepiandrosterone sulfate (DHEAS) levels can be used as a proxy for ACTH, since DHEAS secretion is ACTH-dependent; levels below the reference range and particularly below 40 mcg/dL (1.1 mcmol/L) imply ACTH-independent hypercortisolism.
In ACTH-independent Cushing syndrome, CT of the adrenals usually detects a mass lesion, which is most often an adrenal adenoma. Adrenocortical carcinomas can usually be distinguished from benign adrenal adenomas since they are generally larger (average 11 cm) and many have metastases that are visible on preoperative scans. Adrenal carcinoma is suspected in the following circumstances: (1) diameter of 4 cm or more (average 11 cm diameter); (2) nodule growth; or (3) atypical imaging: density on noncontrast CT greater than 10 Hounsfield units (HU) or CT contrast washout 60% or more or relative contrast washout 40% or more at 15 minutes after intravenous administration.
In ACTH-dependent Cushing syndrome, MRI of the pituitary demonstrates a pituitary lesion in about 50% of cases. Premature cerebral atrophy is often noted. When the pituitary MRI is normal or shows a tiny (less than 5 mm) irregularity that may be incidental, selective catheterization of the inferior petrosal sinus veins draining the pituitary is performed. ACTH levels in the inferior petrosal sinus that are more than twice the simultaneous peripheral venous ACTH levels are indicative of pituitary Cushing disease. Inferior petrosal sinus sampling is also done during ovine CRH (oCRH or Acthrel) administration, which ordinarily causes the ACTH levels in the inferior petrosal sinus to be over three times the peripheral ACTH level when the pituitary is the source of ACTH.
When inferior petrosal sinus ACTH concentrations are not above the requisite levels, a search for an ectopic source of ACTH is undertaken. Location of ectopic sources of ACTH begins with CT scanning of the chest and abdomen, with special attention to the lungs (for carcinoid or small cell carcinomas), the thymus, the pancreas, and the adrenals. In patients with ACTH-dependent Cushing syndrome, chest masses should not be assumed to be the source of ACTH, since opportunistic infections are common. It is prudent to biopsy a chest mass to confirm the pathologic diagnosis prior to resection.
For Cushing syndrome due to ectopic ACTH, CT scanning fails to detect the source of ACTH in about 34% of cases. In such cases, the most sensitive (82%) scanning technique is whole-body imaging with 68Ga-somatostatin receptor-PET/CT (68Ga-DOTATOTATE-PET/CT). The next most sensitive (58%) scanning technique is whole-body imaging with 18F-DOPA-PET/CT.