Cushing's Syndrome

UCLA Endocrinology

Cushing's Syndrome

Mark Goodarzi, M.D.

	Cortisol stimulates the catabolism of peripheral fat and protein to provide substrates for hepatic gluconeogenesis. It also has antiinflammatory effects and modulates the response to stress, dampening defense mechanisms, preventing their dangerous overactivity. It exerts negative feedback on CRH and ACTH. CRH and AVP stimulate ACTH secretion. ACTH is synthesized as a large precursor, pro-opiomelanocortin (POMC), which is processed to ACTH, MSH, b -lipotropin, and b -endorphin which are secreted together.
	Most common cause of Cushing’s syndrome (CS) is exogenous administration of glucocorticoids for chronic inflammatory conditions or after organ transplantation. Also consider inhaled or topical corticosteroids (or mouth rinses containing glucocorticoids, e.g. Klack’s solution). This review will concern endogenous CS.
	Estrogens increase cortisol binding globulin 2-3 fold, which increase total cortisol (90% bound to CBG), which can falsely suggest CS.
	Age and sex may provide diagnostic clues: adrenal carcinoma is more common in children; Cushing’s disease is most frequently seen in women (8:1 compared to men) of child-bearing age (20-40); ectopic CS mostly found in adult males (age 40-60).

PATHOPHYSIOLOGY OF CUSHING’S SYNDROME

I. Corticotropin-dependent (about 80% of cases of CS)

Cushing’s disease: Most common form of CS [68%]: Excessive ACTH secretion by pituitary corticotroph tumors, usually, microadenomas (<1 cm) and most (73-82%) located in lateral portion of pituitary. Macroadenomas are rare (~10% of CD), and corticotroph hyperplasia (increased cell number but preserved architecture, unlike adenomas where reticulin and acinar pattern are lost) and carcinoma (e.g. lymph node metastasis) are extremely rare. See suppressed CRH secretion (by high cortisol) and bilateral adrenal hyperplasia (diffuse, or, rarely macronodular). Usually see exaggerated plasma ACTH and cortisol responses to CRH and incomplete suppression of ACTH and cortisol by dexamethasone. Smaller tumors tend to be more densely granulated and more hormonally active, resulting in more severe symptoms than with large adenomas.

Ectopic ACTH [12-15% of CS]: May present acutely (unique clinical features, see below), most often associated with small-cell lung carcinoma, which accounts for 50-75% of ectopic ACTH, or chronically (indistinguishable clinically from Cushing’s disease but much less common), most often seen with indolent tumors, usually bronchial carcinoids but also thymic, pancreatic carcinoids, medullary thyroid carcinoma, pheochromocytoma, islet cell carcinoma or other neuroendocrine tumors [including pulmonary tumorlets, microscopic nests of neuroendocrine cells]. Ectopic ACTH has been rarely seen in renal cell and breast CA. Ectopic ACTH secretion is usually not suppressed by glucocorticoids. However, ACTH secretion from ~50% of bronchial carcinoids may be suppressed by high doses of dexamethasone.

-Pancreatic endocrine tumors (4-16% of ectopic ACTH) secreting ACTH tend to be aggressive, especially when associated with ZES, and metastasize to regional nodes, liver, kidney, thyroid, peritoneum, and bone. These tumors often produce other hormones (e.g. gastrin).

	Ectopic CRH: [<1% of CS]: Very rare, usually by bronchial carcinoids, clinically indistinguishable from ectopic ACTH. High dose dexamethasone usually able to suppress ACTH secretion, except when tumor also secretes ACTH.
	Pseudo-Cushing’s syndrome

-Major depression [1%]: up to 80% dysregulated cortisol secretion; cortisol hypersecretion usually minimal. HPA axis hyperactivity returns to normal with remission of depression. See normal diurnal cortisol variation.

-Chronic alcoholism [<1%]: Due either to increased CRH or impaired hepatic metabolism of cortisol. Resolves with abstinence (4 weeks) and return of liver function to normal.

-Obesity: 1 mg dexamethasone suppression is falsely positive in up to 30% of obese patients. Urinary free cortisol is positive in up to 50%.

-Severe psychological or physical stress (e.g. HIV infection, sepsis, trauma, eating disorder, sleep apnea, OCD, CNS active drug withdrawal, panic attacks, regular strenuous exercise, last trimester of pregnancy).

II. Corticotropin-independent

	Adrenocortical tumors: 18% of CS: 10% by benign adrenal adenomas, 8% by carcinomas. Carcinomas may occur as part of Li-Fraumeni syndrome or Beckwith Wiedemann syndrome). There is no evidence that such tumors arise from chronic ACTH hypersecretion. Carcinomas are relatively inefficient at synthesizing cortisol, so overproduction of androgenic precursors causing virilization is common. Adenomas synthesize cortisol efficiently, so clinical manifestations are mainly of cortisol excess.
	Bilateral micronodular hyperplasia [1%]: Half of cases occur spontaneously in children and young adults (age < 30); other half occur as primary pigmented micronodular dysplasia (PPNAD), an autosomal dominant disorder associated with the Carney complex: blue nevi, pigmented lentigines, cutaneous, mammary, and atrial myxomas (which can cause stroke and death), pituitary somatotroph adenomas, and testicular tumors, ovarian tumors, cysts, or carcinoma, thyroid nodules or diffuse cancer (>80% of cases have involvement of multiple endocrine systems). 40% of cases are associated with mutant regulatory subunit 1 of PKA (tumor suppressor activity). These patients often respond to dexamethasone with a paradoxical increase in cortisol secretion.
	Bilateral macronodular hyperplasia [<1%]: Very rare; 24 such patients found to have marked cortisol spikes in response to meals with adrenocortical stimulation by gastric inhibitory peptide (can treat with octreotide to decrease GIP). A case of bilateral macronodular hyperplasia (AIMAH) with Cushing’s syndrome (manifest transiently during the patient’s pregnancies and sustained post-menopausally) caused by LH receptor activation was completely reversed by monthly leuprolide (Lupron) injections. Macronodular hyperplasia may occur in McCune Albright syndrome.
	Several other cases of corticotropin-independent Cushing’s (usually AIMAH, rarely adenoma, never carcinoma) have been caused by abnormal adrenal expression and function of receptors for various other hormones: > 20 cases with V1a vasopressin receptor, responsive to posture; b -adrenergic receptor, responsive to isoproterenol and treated with propranolol; interleukin-1, 5-HT₄-receptor.
	Factitious glucocorticoid administration.

CLINICAL FEATURES OF CUSHING’S SYNDROME

Unfortunately non-specific and overlap with much more common disorders such as simple obesity, hypertension, type 2 DM, depression, polycystic ovary syndrome, and syndrome X. Prevalence of CS may be as high as 4% of obese type 2 diabetics with poor sugar control. Typical signs and symptoms:

Weight gain (90%)	Relatively acute, usually central, but can be general. An enlarged dorsocervical fat pad (buffalo hump) accompanies major weight gain of any cause; increased fat pads that fill the supraclavicular fossae are more specific for CS. See thinning of the extremities.
Moon facies	Thickening of facial fat, which rounds the facial contour
Hypertension (85%)	New-onset hypertension in particular. If long-standing, may remain even after hypercortisolism corrected.
Glucose intolerance (80%)	Ranging from hyperglycemia to diabetes
Plethoric facies (80%)	Florid complexion due to telangiectasias
Purple striae (65%)	Violaceous striae wider than 1 cm on abdomen or proximal extremities
Hirsutism (65%)	With acne, usually mild.
Menstrual dysfunction	Oligomenorrhea or amenorrhea; hypercortisolemia causes 2° hypogonadism
Muscle weakness (60%)	With wasting (type II fiber atrophy), proximal weakness manifested by difficulty in climbing stairs, arising from a low chair or squatting.
Easy bruising (40%)	With spontaneous ecchymoses
Osteoporosis (40%)	Trabecular bone loss is greater then cortical, axial greater than peripheral. Thus, vertebral and rib fractures are more common.
Thinning of the skin	Thinning of the skin is more common in older patients or those with chronic CS.
Mental changes	Major depression (most common), insomnia, psychosis, mania, emotional lability
Hematologic	Leukocytosis: neutrophilia (increased release from bone marrow, decreased efflux from circulation, impaired phagocytosis), lymphopenia, eosinopenia, monocytopenia, impaired cell-mediated immunity
Hyperpigmentation	Unusual, except in ectopic ACTH where the ACTH concentration is markedly elevated. Hyperpigmentation does not occur in adrenal tumors.
Hypokalemia	Rare, except in cases of ectopic ACTH or adrenal carcinoma
Other:	Poor wound healing; peripheral edema; decreased libido; increased susceptibility to infection, sometimes life-threatening; spinal epidural lipomatosis (rare, excess epidural fat which can compress cord or nerve roots, causing neurologic deficits); hyperprolactinemia

	Acute Cushing’s syndrome: usually ectopic ACTH from small-cell lung CA: rapid onset hypertension, edema, hypokalemia, alkalosis and glucose intolerance, without enough time to develop typical signs such as moon facies or striae. Mineralocorticoid effects are due to overwhelmed local renal conversion of cortisol (activates aldosterone receptor) to cortisone (does not activate aldo-R). Since it sometimes causes only mild hypokalemia and occurs in patients with rapidly progressive cancer, it is the most underdiagnosed form of CS. CS is a poor prognostic factor in small-cell lung CA (median survival 5.5 months vs. 12 months)
	Excess glucocorticoids cause hypertension by increasing vascular sensitivity to endogenous vasoconstrictors (epinephrine, angiotensin II), increasing cardiac output, activating the renin-angiotensin system by increasing hepatic production of angiotensinogen, decreasing synthesis of vasodilatory prostaglandins. Very high ACTH as in ectopic ACTH production may lead to high cortisol levels which stimulate the mineralocorticoid receptor and may also cause excess production of 11-deoxycorticosterone by the zona fasciculata, leading to mineralocorticoid hypertension.
	Adrenal carcinoma: severe hirsutism and virilization (clitoromegaly, temporal balding). 5-year survival is 20%. May see elevated 17-ketosteroids (urinary androgens; mainly DHEA, DHEA-S and their metabolites) or other cortisol precursors (e.g. DHEA, 11-deoxycortisol) in serum. May see hypoglycemia due to secretion of IGF-2.
	Unrecognized or untreated disease has a 50% 5-year mortality, mainly due to cardiovascular disease.
	Modern surgical and medical therapy for Cushing’s disease has improved survival to essentially the same as the general population (99% 5-year survival, 93% 10-year survival).

DIAGNOSIS OF CUSHING’S SYNDROME

I. Screening tests, tests which detect or confirm suspected hypercortisolism. A normal result on either essentially excludes CS.

Daily 24-hour urinary free cortisol (UFC): Most reliable test because cortisol levels may rise and fall during the day, even in CS. 24-hour excretion gives an integrated daily value. Should do 2-3 collections since often performed incorrectly (excretion of creatinine (males: 20-25 mg/kg/d, females 15 mg/kg/d) should not vary by >10% between collections) or falsely negative due to daily fluctuation in cortisol levels (10-15% of patients with CS have 1 of 4 24-hr UFC in the normal range) or abnormalities caused by other medications taken by the patient. If UFC < 90 µg/d (on all 3 samples), CS is unlikely; if > 250-300 µg/d, CS is present; 90-300 µg/d warrants testing to distinguish CS from pseudo-CS. Renal failure can produce a false-negative test.

-HPLC: High performance liquid chromatography, recently introduced, is more specific (less false elevation) than RIA (which picks up other glucocorticoids), however drug metabolites (e.g. carbamazepine) can cause spuriously elevated values. For HPLC, normal range is < 50 µg/d.

-Increased fluid intake, by increasing urine volume (> 4 L/d) and thus reducing the fraction of filtered cortisol that is metabolized or reabsorbed, can increase UFC measurements.

Low-dose dexamethasone suppression test (L-DST): dexamethasone, a synthetic, potent corticosteroid, is given to suppress ACTH. Test is very sensitive*, but lacks specificity: false positives in 12-30% (many pseudo-CS), thus, must confirm any positive test with a 24-hour UFC.

a. Two-day, low-dose DST (0.5 mg every six hours for eight doses), with 24-hr UFC collection during second day. UFC > 10 µg/d (or urinary 17-hydroxycorticosteroid (17-OHCS, mainly metabolites of cortisol, cortisone, 11-deoxycortisol) > 2.5 mg/d) indicate CS.

b. Overnight DST (1 mg at 11 p.m. or midnight), with 8 a.m. plasma cortisol the next morning. Normally a.m. cortisol will suppress to < 5 µg/dl. A.M. cortisol > 10 µg/dl indicates CS.

-Measuring plasma dexamethasone can clarify confusing results, caused by noncompliance, individual variability in dexamethasone metabolism, or the effects of drugs on steroid metabolism (e.g. CYP3A4 inducers phenytoin, barbiturates, rifampin, dex itself, primidone, troglitazone increase dexamethasone metabolism, leading to false positives). CYP3A4 inducers do not affect the 50 mg hydrocortisone suppression test (normal response corticosterone < 120 ng/dL or 50% baseline).

-17-OHCS measurements are less accurate than UFC since metabolite excretion is variable depending on body weight.

*Recent studies have found cases (up to 18%) of Cushing’s disease which were unusually sensitive to dexamethasone and suppressed on L-DST, making L-DST a test with poor sensitivity. A new hydrocortisone suppression test has been proposed (less potent than dexamethasone).

II. Tests to distinguish Cushing’s from pseudo-Cushing’s

	Diurnal variation: Normally, cortisol levels peak at 4-8 a.m., and decline through the day to a nadir (< 50-80% of a.m. values) at midnight to 2 a.m. Patients with true CS lose this diurnal variation, whereas it is preserved in pseudo-Cushing’s. A midnight cortisol > 7.5 µg/dl indicates CS, while a value < 5 µg/dl rules it out. Must be drawn from an indwelling catheter after a 3 hour fast and 2 hours of rest.
	Dexamethasone-CRH test: This test exploits the greater cortisol response to CRH in Cushing’s disease and the relatively more suppressible ACTH secretion in pseudo-Cushing’s. Dexamethasone (0.5 mg q6 hours x 8 doses, 1st dose at noon, last at 6 a.m.) and CRH (1 µg/kg IV on 3rd morning, 8 a.m.) are given in sequence. In pseudo-Cushing’s, plasma cortisol at 15 minutes is low after dex and remains low after CRH, whereas in CS, cortisol is not as low after dex and increases (> 1.4 µg/dl) after CRH is given. Need very good cortisol assay.
	Two-day, low-dose DST is used by some to exclude pseudo-Cushing’s. In pseudo-Cushing’s, usually see suppression of cortisol < 3 µg/dl and 24-hour UFC < 20 µg/d.
	Salivary cortisol. Avoids the stress of venipuncture and allows outpatient testing, including midnight sampling. Reflects free cortisol due to absence of binding proteins in saliva. May be used to demonstrate lack of diurnal variation or lack of suppression with dexamethasone. May be particularly useful in cyclic CS since it allows repeated outpatient sampling. An 11 p.m. salivary cortisol > 3.6 nmol/L is 92% sensitive for CS.
	CRH levels (expect to be suppressed in CS): Useless due to low levels of CRH in serum and extrahypothalamic production of CRH.

III. Tests to distinguish corticotropin-dependent and corticotropin-independent.

	ACTH: ideal time to measure plasma ACTH and cortisol is between midnight and 2 a.m., when concentrations of these are normally at their lowest; however, late afternoon (4 p.m. or later) assays are usually acceptable. Since their secretion is episodic, should obtain 2-3 measurements. If cortisol > 15 µg/dl and ACTH < 5 pg/ml, cortisol secretion is ACTH-independent. If ACTH > 15 pg/ml, the cortisol secretion is ACTH-dependent. Intermediate ACTH is less definitive but usually indicate ACTH-dependence. A profoundly elevated ACTH > 250 pg/ml suggests ectopic ACTH production. ACTH levels measured by IRMA has greater specificity and sensitivity than RIA; however, IRMA cannot detect an unusual type of ACTH ("big" corticotropin), which may be biologically active.
	CRH stimulation test (done early morning, fasting): Some experts claim that low ACTH levels may be misleading due to episodic and pulsatile secretion from tumors, and thus suggest a low ACTH (< 10 pg/ml) be followed by a CRH stimulation test (100 µg IV, ACTH measured 15 & 30 min. later): peak ACTH < 10 pg/ml indicates ACTH-independent disease and should prompt adrenal imaging.
	DHEA-S: measured when plasma cortisol is not suppressible, may suggest adrenal adenoma if < 0.4 µg/ml. DHEA-S will be elevated in adrenal carcinoma.

IV. Tests to determine the source of excess corticotropin secretion. Since most (~90%) have Cushing’s disease, the goal is to detect the few patients with ectopic ACTH. Unfortunately, noninvasive tests (H-DST, metyrapone) have only 60-90% sensitivity when cortisol cutoffs are set at a level to provide 100% specificity for identifying ectopic ACTH.

High-dose DST. Exploits the fact that in Cushing’s disease there is relative resistance to glucocorticoid negative feedback while in ectopic ACTH (and adrenal Cushing’s), there is usually complete resistance. However, ~15% of Cushing’s disease cases do not suppress. Thus, a test which suppresses is more useful (UFC suppression < 10% is 97% specific for CD). However, some carcinoids secreting ACTH suppress on H-DST. Of those who do not suppress on H-DST, 60-70% have Cushing’s disease.

a. Two-day, H-DST (2 mg every six hours for eight doses) with measurement of 24-hour UFC and 17-OH-corticosteroids. Test often done sequentially after L-DST (Liddle test). Cushing’s disease is indicated by > 90% suppression of UFC or > 64% suppression of 17-OHCS compared to baseline values. High non-suppressible 17-OHCS suggest adrenal carcinoma. An increase in UFC is seen in most cases of primary pigmented micronodular dysplasia.

b. Overnight H-DST (8 mg given at 11 p.m. to midnight): Measure baseline a.m. cortisol and ACTH and 8 a.m. next day. Reported to have greater sensitivity than two-day test. In Cushing’s disease, follow-up values usually decrease by 50% from baseline.

	Metyrapone stimulation test: Metyrapone blocks conversion of 11-deoxycortisol to cortisol; the fall in cortisol normally stimulates ACTH from the pituitary, leading to increased 11-deoxycortisol and urinary 17-OHCS (includes metabolites of 11-deoxycortisol). 750 mg given every 4 hours for 6 doses. Patients with Cushing’s disease have a normal or supranormal increase in plasma 11-deoxycortisol (400-fold) or urinary 17-OHCS (increase > 70% over baseline), whereas most patients with ectopic ACTH (or adrenal CS) have little or no increase in either because pituitary ACTH is suppressed.
	CRH stimulation test: Controversial value in distinguishing pituitary vs. ectopic ACTH. Expect greater increase in ACTH level (> 35-50% baseline) after CRH administration in Cushing’s disease; however, there is much overlap of peripheral plasma ACTH responses.
	Inferior petrosal (venous) sinus sampling (IPSS). The most direct and accurate (sensitivity and specificity approach 100%, if done by experienced team) method to demonstrate pituitary hypersecretion of ACTH. Must be done when patient is hypercortisolemic (so that normal corticotroph cell secretion is suppressed). The petrosal sinuses drain the pituitary via the cavernous sinuses. Catheters are passed via bilateral femoral veins, through the internal jugulars to both inferior petrosal sinuses, and peripheral and IPS ACTH are measured before and after CRH stimulation (100 µg or 1 µg/kg of ovine CRH). Cushing’s disease is diagnosed if basal IPS:peripheral ACTH > 2 or post-CRH IPS:peripheral ACTH > 3. Procedure is invasive, and must be done at a specialized center. Catheters must be placed high enough to avoid drainage from the contralateral side. Both sides must be sampled because a gradient is sometimes detected only on the side with the adenoma. IPSS can be misleading in patients with anomalous venous drainage. Complications include poor catheter placement, inguinal hematomas, cavernous sinus thrombosis, infection, hemorrhage, and brain stem infarction. An interpetrosal gradient (e.g. right vs. left) > 1.4 after CRH predicts location of the lesion in 60-75% of patients. Formerly viewed as a test done late in the workup, it is increasingly being advocated as an early test due to its greater accuracy and additional helpfulness in localizing the side of the pituitary harboring the adenoma. By avoiding extensive biochemical testing which often requires hospitalization and preventing unnecessary pituitary surgery, proceeding directly to IPSS may actually be more cost-effective.
	Cavernous sinus sampling (CSS): similar to IPSS except catheter advanced further into cavernous sinus. Has extra benefit of providing more accurate (83%) information about intra-pituitary location of adenoma.
	High jugular vein sampling: Central to peripheral ACTH gradients are diagnostic even when obtained from proximal jugular vein. Less invasive test, but only useful if a gradient is present (if not, must do IPSS) and provides no information on adenoma lateralization. As with CSS, not widely used.
	Some experts claim that, once ACTH-dependence is established, the specificity if MRI is sufficient that if MRI demonstrates a lesion, it is reasonable to proceed directly to transsphenoidal surgery. Others urge IPSS be done first.

V. Imaging procedures: Must be guided by biochemical tests, since imaging provides no information about function. Use only to locate a tumor.

	If lab workup suggests ACTH-independent Cushing’s, thin-section CT (CT preferred to MRI, however, higher T2 intensity on MRI indicates carcinoma rather than adenoma) of adrenal glands is usually the final diagnostic maneuver. Caution: 2-15% of patients have nonfunctioning adenomas (incidentalomas). Patients with primary pigmented micronodular dysplasia classically have glands with an irregular contour (beads on a string), but may have normal-appearing glands or macronodules. Note: With ectopic ACTH, adrenal enlargement can be marked, primarily as a result of hyperplasia of the zona reticularis.
	Iodo-seleno-cholesterol scans are used to evaluate synthetic function within the adrenal gland. May be needed for small adrenal tumors.
	Chest CT or MRI: most ectopic ACTH secreting tumors reside in the chest (small cell lung CA, bronchial or thymic carcinoids). MRI better at detecting mediastinal lesions; however, CT scan is preferred. Abdominal imaging rarely detects occult ACTH secreting tumors and should be done only if chest imaging is negative (at which point biochemical evaluation for pheochromocytoma and medullary carcinoma of the thyroid should be done).
	Radionuclide imaging with indium-111 labeled octreotide detects up to 86% of carcinoid tumors, many of which have somatostatin receptors. May also detect pheochromocytoma.
	Pituitary MRI (with and without gadolinium; more sensitive than CT) only ~50% sensitive in detecting corticotroph microadenomas. When positive, MRI does help localize tumor (false lateralization in 4-14%). Even the presence of a pituitary lesion on MRI, especially if size < 4 mm, does not necessarily confirm functionality because 10-15% of the general population harbor pituitary incidentalomas. This limits MRI specificity to 90%. IPSS has been found to be more accurate than MRI in localization of the pituitary lesion.
	Pituitary intraoperative ultrasound using a sterile probe once bony wall of sella is removed: 76% sensitive (tumors appear hyperechoic).

VI. Novel tests

Serum cortisol at 0800 and 1600 at baseline (day 1). Dexamethasone 0.5 mg po q6hr is taken day 2 and 3, serum cortisol measured 1600 day 3 (low dose dexamethasone). 2 mg q6hr is taken day 4 and 5, serum cortisol obtained 1600 day 5 (high dose); serum DHEA-S obtained 0800 day 1 (baseline). Nonsuppressible values defined as cortisol > 5 µg/dl on day 3 and > 10 µg/dl on day 5. In the presence of nonsuppressible cortisol, DHEA-S < 0.4 µg/ml suggests adrenal adenoma. This method can be done on outpatients and avoids the inconvenience of urine collection.

AVP V3 receptors have been found in corticotroph adenomas. Increases in cortisol (>20%) and ACTH (>50%) after IV DDAVP (5-10 µg) suggest Cushing's disease is present. However, ectopic ACTH (bronchial carcinoid) may also respond. DDAVP may prove useful in monitoring Cushing's disease patients after surgery (no response if cured) and to rule out pseudo-Cushing's (no response).

VII. Diagnostic dilemmas

	Ectopic CRH will produce a central to peripheral ACTH gradient on IPSS and usually leads to surgery which shows corticotroph hyperplasia. Thus, all patients undergoing IPSS should have CRH measurements.
	An entity of cyclic Cushing’s syndrome exists where hypercortisolemia is episodic. Thus, biochemical testing will be negative if done during a period of eucortisolemia.

Crooke’s cell adenoma. Crooke’s cells are corticotrophs which have been chronically suppressed by cortisol excess. These cells have cortisol receptors and show hyaline change. Some adenomas have a component of Crooke’s cells, and exhibit variable suppressibility by dex. Perhaps these are responsible for episodic Cushing's?

Challenging cases which have been encountered in a large series: corticotroph cell hyperplasia not dex suppressible; corticotroph hyperplasia in varying patterns (nodular, diffuse, multifocal cell nests, "adenomatous hyperplasia" [a discrete mass of corticotrophs still organized in acinar structure]); IPSS positive Cushing's caused by a pituitary adenoma located in the sphenoid sinus; a case of Nelson's syndrome co-existing with Cushing's (due to adrenal remnant); sequential disease states; dual or collision tumors; decoy or incidental lesions; absence of pathologic diagnosis.

TREATMENT OF CUSHING’S SYNDROME

I. Cushing’s disease

	Transsphenoidal microadenomectomy: treatment of choice, if a clear lesion has been identified and is resectable. Otherwise, 85-90% resection of the anterior pituitary should be performed. A hemi-hypophysectomy can be done if lateralization information is available and avoids hypopituitarism. Cure rate is about 85% (up to 93% for isolated tumors; <50% for invasive tumors; 90% for microadenomas (< 1 cm), 65% for macroadenomas), with less success for repeat surgery. One large series (185 cases) had an 18% overall recurrence rate. Criteria for cure include undetectable plasma cortisol (< 1 µg/dl) the day after surgery and a plasma ACTH < 5 pg/ml 24 hours after the last dose of hydrocortisone, 4-7 days after surgery.
	Complications (~5%): diabetes insipidus (usually transient), CSF rhinorrhea, meningitis, sinusistis, hemorrhage. Hypercortisolism recurs in only 5%, with more recurrence with macroadenomas (relative risk 5.1).
	Patients require daily glucocorticoid-replacement from the time of surgery until their HPA axis recovers, usually 6-12 months after surgery.
	Pituitary irradiation: may be performed instead of microadenomectomy if fertility is desired. Irradiation is indicated for patients not cured by surgery, and corrects hypercortisolism in ~45% of adults (cures 85% of children and may thus be used as initial therapy). Stereotactic radiation (e.g. gamma knife, proton beam) may prove more useful and more convenient (single day therapy vs. 6 week course with conventional). Three to 12 months are needed for full benefit from irradiation, during which time adrenal enzyme inhibitors may be used. Irradiation also reduces the incidence of Nelson’s syndrome (enlarged, locally invasive, ACTH-secreting pituitary tumors occurring after bilateral adrenalectomy that usually develop in patients who presented with invasive tumors).
	If the above fail, bilateral total adrenalectomy is curative (will need lifelong glucocorticoid and mineralocorticoid replacement).

II. Ectopic corticotropin and CRH syndromes

	Resection of nonpituitary tumors is curative, but this is usually not possible. Hypercortisolism may be controlled with adrenal enzyme inhibitors or mifepristone.
	If the patient has an indolent, nonresectable tumor (relatively longer life expectancy), bilateral adrenalectomy (surgical or medical) will cure hypercortisolism.

III. Primary adrenal disease

	Bilateral micro- or macronodular hyperplasia is cured by bilateral total adrenalectomy.
	Unilateral adrenalectomy cures adrenal adenoma. Since the contralateral adrenal gland is suppressed, glucocorticoid replacement is needed until adrenal function returns (up to 15 months).
	Adrenal carcinoma is also treated with unilateral adrenalectomy, but usually (75%) has recurrence which does not respond to radiation or chemotherapy. Mitotane (4 g per day) given after initial surgery may prevent recurrence.
	Aggressive surgical treatment of recurrent or metastatic disease may prolong life.

MEDICAL THERAPY FOR CUSHING’S SYNDROME

I. Adrenal enzyme inhibitors are useful adjunctive therapy to control acute symptoms of hypercortisolemia, during treatment with irradiation, in preparation for surgery or whenever definitive treatment is delayed. They are effective in the majority of cases in a dose-dependent manner. However, these medications do not cure the condition.

	Ketoconazole: Interferes with fungal ergosterol synthesis and mammalian cholesterol synthesis, also inhibits several cytochrome P450 enzymes (C17-20 lyase mainly [more suppressive of testosterone than cortisol], also cholesterol side chain cleavage, and 11ß/18-hydroxylation) first choice since few side effects. Cortisol secretory rates can be reduced to near-normal levels. Dose range 200-1000 mg bid; usually 600-800 mg/day total is required to normalize UFC (high normal, to avoid risk of adrenal insufficiency). Most common side effects are GI reaction, pruritus, and hepatotoxicity (idiosyncratic), so LFT’s must be monitored. At high doses, gynecomastia, impaired testicular function, and adrenal insufficiency may occur. Since it blocks early in the steroid pathway, there is no accumulation of potentially toxic steroids. Patients can be maintained on ketoconazole for years. Therapy can be combined with other agents (aminoglutethimide).
	Metyrapone: 11ß-hydroxylase inhibitor (also inhibits 18-hydroxylase, and cholesterol SCC), average dose 250-750 mg tid-qid, can normalize cortisol levels. Can lead to increases in androgens (acne, hirsutism) and in deoxycorticosterone, which has mineralocorticoid activity and can lead to hypertension and hypokalemia.
	Aminoglutethimide: A potent inhibitor of adrenal steroidogenesis (inhibits cholesterol SCC, 11ß/18-hydroxylation), it is started at 250 mg qid, up to 2 g daily. Must monitor for hypoadrenalism (plasma cortisol, UFC) and hypoaldosteronism. Also is a potent inducer of hepatic metabolizing enzymes (drug interactions) and enhances its own metabolism, which accounts for development of tolerance to early side effects (lethargy, dizziness, ataxia, rashes). May also cause hypothyroidism and hematologic toxicity. More effective in adrenal CS or ectopic ACTH than Cushing’s disease. May be combined with metyrapone.
	Mitotane: Mechanisms of action: adrenocorticolytic effects (medical adrenalectomy), modification of peripheral steroid metabolism, direct inhibition of steroid biosynthesis (inhibits cholesterol SCC and 11ß/18-hydroxylase activities). Mitotane is palliative for recurrent or residual adrenal carcinoma, starting at 250 mg qid and increasing to tolerance levels (24 g/day). In Cushing’s disease daily doses of 2-4 g are used, and is more effective when combined with pituitary irradiation. Extensive metabolism of mitotane is required for action, and it takes several weeks to reduce cortisol secretion. It often causes anorexia, nausea, vomiting, diarrhea, hypercholesterolemia, and CNS effects such as somnolence, dizziness, vertigo, and ataxia. Patients on this drug often need increased doses of glucocorticoid and mineralocorticoid replacement therapy and should be monitored for hypoadrenalism.

II. Compound with peripheral receptor site of action.

Mifepristone (RU486): A potent glucocorticoid receptor antagonist. Limited experience in CS. Drawback is the absence of biochemical markers to monitor clinical response and avoid adrenal insufficiency. Nausea, drowsiness, hypoadrenalism and gynecomastia have been reported.

III. Compounds with hypothalamic-pituitary site of action. Neuromodulatory agents influence ACTH secretion because ACTH hypersecretion remains under hypothalamic control even in Cushing’s disease. These agents are generally ineffective as sole therapy.

	Serotonin antagonists (cyproheptadine, ritanserin): Central serotonergic pathways may facilitate pituitary ACTH release. A few patients with Cushing’s disease have responded with normalization of ACTH and cortisol levels.
	Dopamine agonist (bromocriptine): Limited effectiveness in controlling ACTH hypersecretion in Cushing’s disease.
	GABA agonist (sodium valproate): Limited effectiveness in suppressing ACTH secretion in Cushing’s disease by decreasing hypothalamic CRH release.
	Somatostatin analogs (octreotide): Ineffective in Cushing’s disease, but has controlled ACTH hypersecretion and controlled tumor growth in Nelson’s syndrome (pentetreotide-scan positive). Somatostatin analogs may also suppress ACTH secretion by carcinoids, which may have somatostatin receptors.

References (v1.3)

Ashcraft M, Van Herle AJ, et al. Serum cortisol levels in Cushing's syndrome after low- and high-dose dexamethasone suppression. Ann Int Med 1982;97:21-26.
Brandt JL, Mark EJ. A 64-year-old man with Cushing's syndrome and a pancreatic mass. N Engl J Med 2000;342:414-20.
Cook DM, Loriaux DL. Cushing’s syndrome. Curr Ther Endo Met 1994;5:142-6.
Findling JW, Raff H. Newer diagnostic techniques and problems in Cushing’s disease. Endo Met Clin N Am 1999;28:191-210.
Graham KE, Samuels MH. Recent advances in the evaluation of Cushing’s syndrome. Endocrinologist 1998;8:425-35.
Hasinski S. Assessment of adrenal glucocorticoid function. Postgrad Med 1998;104:61-71.
Koch CA, Doppman JL, et al. Spinal epidural lipomatosis in a patient with the ectopic corticotropin syndrome. N Engl J Med 1999;341:1399-1400.
Lacroix A, Hamet P, Boutin JP. Leuprolide acetate therapy in luteinizing-hormone dependent Cushing’s syndrome. N Engl J Med 1999;341:1577-81.
Orth DN. Cushing’s syndrome. N Engl J Med 1995;332:791-803.
Smyth, Harley. Lecture: Clinical, biochemical and pathological correlations in Cushing’s disease: a histological and longitudinal analysis of 185 operated cases. June 7, 2000.
Sonino N, Boscaro M. Medical therapy for Cushing’s disease. Endo Met Clin N Am 1999;28;211-22.
Stern N, Tuck M. The adrenal cortex and mineralocorticoid hypertension. In: Lavin N, ed. Manual of endocrinology and metabolism. Boston: Little, Brown and Co., 1994:111-29.
Stratakis CA, Sarlis N, et al. Paradoxical response to dexamethasone in the diagnosis of primary pigmented nodular adrenocortical disease. Ann Intern Med 1999;131:585-91.
Swearingen B, Biller BMK, et al. Long-term mortality after transsphenoidal surgery for Cushing’s disease. Ann Intern Med 1999;130:821-4.

[ Home ] [ Up ]