The mechanism whereby the POMC gene becomes derepressed in noncorticotroph tumors is not understood. One hypothesis is that these cells are derived from a common multipotential progenitor cell capable of producing peptide hormones, such that ACTH production is a reversion to a less differentiated state.22 The speculation that many ACTH-producing tumors are derived from neural crest amine precursor uptake and decarboxylation (APUD) cells may support this view,23 although this embryological hypothesis is not supported by the most recent data. However, because endodermally derived tumors also produce ACTH, the acquisition of APUD characteristics may be but one manifestation of dedifferentiation and may not represent the cause of ectopic ACTH production.
Although the mechanism of gene derepression is not understood, the regulation of POMC production and processing has been investigated. POMC, corticotropin-like intermediate lobe protein, and larger forms of ACTH (“big” or pro-ACTH) that are not usually secreted may circulate, and the intracellular ratio of the POMC products may be abnormal.24,25 Investigation of cell lines of small cell carcinoma of the lung that synthesize POMC and pro-ACTH showed that only ACTH precursors were secreted, suggesting that processing to ACTH is defective.26 The pattern of POMC mRNA species in ACTH-producing tumors has been characterized. A 1200 bp transcript similar to that of a corticotroph adenoma,27 a shorter than normal 800 bp mRNA lacking a signal sequence for secretion,27,28 and a larger 1400 to 1500 bp POMC transcript have been identified. The larger species appears to originate upstream of the usual pituitary promoter, with preservation of the normal translation start site.29,30 It is possible that the promoters that initiate this transcription are not regulated by glucocorticoids, and this may explain in part the lack of responsiveness to glucocorticoid suppression noted clinically in these patients. In vitro investigation of human small cell cancer cell lines and pancreatic islet cell tumors with normal glucocorticoid receptor binding has found, for the most part, no regulation of POMC, tyrosine aminotransferase, or the glucocorticoid receptor mRNA at doses of hydrocortisone that would normally suppress pituitary production.31–33 However, clinical observation of suppression of ACTH production by some bronchial carcinoids during glucocorticoid administration suggests retention of a functional glucocorticoid response element that regulates POMC production, at least in some ectopic tumors.34
Ectopic Corticotropin-Releasing Hormone Secretion
Tumor secretion of CRH with or without ACTH secretion is a rare cause of Cushing’s syndrome. Although many tumors immunostain for CRH, its secretion is less common, and most patients do not develop cushingoid features.35 Thus, the diagnosis primarily rests on the demonstration of elevated plasma CRH levels (requiring an assay that is not readily available). The literature includes fewer than 20 patients who fit this criterion. Tumors may have negative immunostaining for ACTH, but this may be related to reduced storage and rapid secretion. In cases such as these, a CRH and ACTH gradient across the tumor bed can be suggestive that, in fact, the tumor secretes both peptides.36 Tumors include bronchial and thymic carcinoids, small cell lung cancer, medullary thyroid carcinoma, pheochromocytoma, gangliocytoma, prostate carcinoma, and ganglioneuroblastoma.37,38 The biochemical responses to diagnostic tests can be similar to those seen in ectopic ACTH secretion or in pituitary ACTH-dependent disease.38 It is important to note that many, if not all, ectopic secretors of CRH causing Cushing’s syndrome are also ectopic ACTH secretors.
Primary Adrenal Disease
The primary adrenal forms of Cushing’s syndrome do not share a common cause. Although the cause of adrenocortical neoplasia is not known, some events important in the development of adrenal cancer have been identified. Paternal isodisomy at 11p15.5 with overexpression of insulin-like growth factor-2 (IGF-2) and reduced expression of CDKN1C (a G1 cyclin-dependent kinase inhibitor) and H19 (a putative growth suppressor) seems to be a key event. Mutations of p53 may be involved in a small subset of carcinomas, and mutations of β-catenin may be an early event. Other genes important in pathogenesis remain to be elucidated, although potential loci have been identified at chromosomes 17p, 1p, 2p16, and 11q13 for tumor suppressor genes, and at chromosomes 4, 5, and 12 for oncogenes.39 Adenomas and carcinomas tend to be monoclonal, although the nodular hyperplasias are often polyclonal.40 Adrenal adenomas are encapsulated benign tumors, usually less than 40 g in weight. Adrenal carcinomas usually are encapsulated, generally weigh more than 100 g, and may lack classic histologic features of malignancy, although nuclear pleomorphism, necrosis, mitotic figures, and vascular or lymphatic invasion suggest the diagnosis.41 The adjacent adrenal tissue is atrophic in both conditions.
Primary pigmented nodular adrenal disease (PPNAD), also known as micronodular adrenal disease, is a rare form of Cushing’s syndrome characterized histologically by small to normal-size glands (combined weight <12 g) with cortical micronodules (average 2 to 3 mm) that may be dark or black in color. The intervening cortex is usually atrophic.42 Most cases of PPNAD occur as part of the Carney complex in association with a variety of other abnormalities, including myxomatous masses of the heart, skin, or breast; blue nevi or lentigines; and other endocrine disorders (sexual precocity; Sertoli cell, Leydig cell, or adrenal rest tumors; acromegaly). The Carney complex is inherited as an autosomal dominant condition, and Cushing’s syndrome occurs in approximately 30% of cases. The tumor suppressor gene PRKAR1A, coding for the type 1A regulatory subunit of protein kinase A, has been shown to be mutated in approximately one half of patients with Carney complex. Mutations in this gene and also the phosphodiesterase 11A (PDE11A) gene have been shown to be associated with an isolated distinct form of PPNAD.43
Cushing’s syndrome resulting from bilateral nodular adrenal disease is an uncommon feature of the McCune-Albright syndrome,44 which is characterized by fibrous dysplasia of bone, café-au-lait skin pigmentation, and endocrine dysfunction (usually precocious puberty). In this disease, an activating mutation at codon 201 of the α subunit of the G protein that stimulates cyclic adenosine monophosphate formation occurs in a mosaic pattern in early embryogenesis.45 If this affects some adrenal cells, constitutive activation of adenylate cyclase and the steroidogenic cascade leads to nodule formation and glucocorticoid excess. The internodular adrenal cortex, where the mutation is not present, becomes atrophic.46
A missense mutation of the ACTH receptor, resulting in its constitutive activation and ACTH-independent Cushing’s syndrome, also has been reported.47
ACTH-independent bilateral macronodular adrenal hyperplasia (AIMAH) is a rare form (<1%) of Cushing’s syndrome that involves large or even huge adrenal glands, usually with definite nodules on imaging. Most cases are sporadic, but a few familial cases have been reported.48 Although the cause remains unclear in most cases, some nodules express increased numbers of receptors normally found on the adrenal gland, or ectopic receptors for circulating ligands that then can stimulate cortisol production. Perhaps the best known example of this phenomenon is food-dependent Cushing’s syndrome. The normal postprandial increase in gastric inhibitory peptide (GIP) appeared to cause Cushing’s syndrome in two middle-aged women with bilateral multinodular adrenal enlargement, mildly elevated urinary free cortisol (UFC) values, and undetectable plasma ACTH values. Fasting morning serum cortisol values were low or normal. Cortisol values increased dramatically after meals and after in vivo or in vitro exposure to GIP.7,8 In one patient, curative bilateral adrenalectomy revealed multinodular adrenal glands weighing 20 and 35 g.8 In the other, treatment with octreotide ameliorated the syndrome.7 Ectopic expression of GIP receptors was found in these patients. Aberrant expression of vasopressin, β-adrenergic luteinizing hormone/human chorionic gonadotropin, serotonin, angiotensin, leptin, glucagon, interleukin (IL)-1, and thyroid-stimulating hormone (TSH) has been described as functionally linked to cortisol production.49 However, it is possible that this apparent ectopic induction of receptors on the adrenal is a response to the adrenal hyperplasia rather than its cause.
Adrenal rest tissue in the liver, in the adrenal beds, or in association with the gonads may rarely cause Cushing’s syndrome, usually in the setting of ACTH-dependent disease after adrenalectomy.50–53 Ectopic cortisol production by an ovarian carcinoma has been reported.54
PSEUDO-CUSHING’S STATES
A pseudo-Cushing’s state may be defined as one in which some or all of the clinical features that resemble true Cushing’s syndrome, and some evidence of hypercortisolism, are present but disappear after resolution of the underlying condition.55 The pathophysiology of these states has not been established. One hypothesis is that these stressful conditions increase the activity of the CRH neuron, resulting in excessive ACTH secretion, adrenal hyperplasia, and increased cortisol production.56 The model predicts only intermittent and modest hypercortisolism because of appropriate corticotroph reduction in ACTH secretion in response to negative feedback by cortisol (Fig. 15-1). This construct presumes also that hypertrophied adrenal glands produce excessive glucocorticoids in response to normal ACTH levels, an assumption that is supported by the blunted ACTH, but not cortisol, response to exogenous CRH in anorexia nervosa,57 depression,58 and obligate athleticism.59
FIGURE 15-1. Physiology of the hypothalamic-pituitary-adrenal axis in normal individuals and hypercortisolemic states. Corticotropin-releasing hormone (CRH) secretion from the hypothalamus normally stimulates adrenocorticotropic hormone (ACTH) secretion from the pituitary gland. This in turn results in increased cortisol production from the adrenal glands. The system is modulated by negative feedback inhibition by cortisol of both CRH and ACTH secretion. In pseudo-Cushing’s syndrome, the CRH neuron is activated by central input (large shaded arrow), resulting in increased CRH output that eventuates in hypercortisolism. Increased cortisol production restrains corticotroph activation but does not completely reverse the activation of the CRH neuron, so that mild to moderate hypercortisolism may persist. In Cushing’s disease, a corticotroph adenoma secretes ACTH in excess and is inhibited only partially by rising cortisol levels. In this setting and that of ectopic ACTH secretion and primary adrenal disease, the CRH neuron is suppressed by hypercortisolism. In ectopic ACTH secretion, excessive secretion of ACTH from a nonpituitary tumor is not inhibited by glucocorticoid feedback. In this setting and that of autonomous production of cortisol by the adrenal gland, ACTH secretion by normal corticotrophs is suppressed by hypercortisolism.
EPIDEMIOLOGY
Iatrogenic causes account for most cases of Cushing’s syndrome because of the common therapeutic use of high-dose glucocorticoids. Large series have reported the distribution of endogenous cases as follows: Cushing’s disease (68%), adrenal adenomas (8% to 19%), adrenal carcinoma (6% to 7%), ectopic ACTH syndrome (6% to 15%), and nodular adrenal hyperplasia (2%).55,60 However, a paucity of information is available on the true incidence of these causes. Perhaps the best data come from a population-based study covering the whole of Denmark (population of 5.3 million), which used stringent methods of data collection, aided by the small number of centers treating the disorder.61 The incidences of Cushing’s disease, adrenal adenoma, and adrenal carcinoma were 1.2 per million per year, 0.6 per million per year, and 0.2 per million per year, respectively. The reported incidence of ectopic ACTH syndrome was extremely low (0.1 per million per year). This is probably due (as the authors concede) to the fact that many cases were never recognized, but may be explained in part by a group of patients with ACTH-dependent Cushing’s syndrome (0.5 per million per year) with presumed but unproven pituitary disease. Some of these may well have had ectopic ACTH syndrome. The incidence of ectopic ACTH syndrome most certainly is underestimated in the endocrine literature because most cases reaching endocrinologists are those caused by occult tumors as opposed to those caused by overt malignancy. However, given that Cushing’s syndrome will be present in 3% to 12% of cases of small cell lung cancer,62,63 and that the recent incidence of small cell lung cancer in Europe is approximately 120 per million per year in men and 40 per million per year in women,64 this is by far the most common cause. Other epidemiologic studies have looked at just the incidence of Cushing’s disease and have found rates between 0.7 per million per year in northern Italy65 and 2.4 per million per year in northern Spain.66
Gender and age distribution varies with the cause of Cushing’s syndrome. Adrenal adenomas and Cushing’s disease present much more commonly in women than in men, and adrenal carcinoma is approximately 1.5 times as common as in men.55,60 Nodular adrenal hyperplasia has an approximately equal gender ratio.
Ectopic ACTH syndrome is the only cause of the syndrome that is more common in men (other than Cushing’s disease in prepubertal children), although this may change as more women are developing small cell lung cancer. Lung cancer is more common after age 40, and this accounts for the increased mean age of patients with ectopic ACTH syndrome compared with Cushing’s disease, which occurs between 25 and 40 years of age.67 The other major cause of ectopic ACTH secretion, intrathoracic carcinoids, has a peak incidence around 40 years and only a slightly increased male-to-female ratio.68 The age distribution of adrenal cancer is bimodal, with peaks in childhood and adolescence and late in life, although adrenal adenoma occurs most often around 35 years of age.
CLINICAL FEATURES
Excessive cortisol production has widespread systemic effects67,69–72 (Table 15-3). Although the full-blown cushingoid phenotype is unmistakable, the clinical diagnosis may be equivocal for patients with few of the typical characteristics (Fig. 15-2). Some nonspecific features consistent with the diagnosis of Cushing’s syndrome, such as obesity, hypertension, and menstrual irregularity, are common in the general population and may provoke unwarranted and costly screening tests for patients not likely to be affected.
Table 15-3. Percentage Frequency of Clinical Signs and Symptoms of Cushing’s Syndrome as Described in Six Large Studies from 1952 to 2003
FIGURE 15-2. Body habitus of two patients with proven Cushing’s syndrome. Features typical of the syndrome—central obesity, round face, and supraclavicular fat pads—are present in the patient in A, but not in the patient in B, illustrating that the diagnosis is not always apparent from the initial physical examination.
One useful strategy when the diagnosis of Cushing’s syndrome is considered, is to look for evidence of progressive physical changes by examination of serial photographs, especially of individuals photographed at annual events such as holidays, birthdays, or school milestones (Fig. 15-3). Another approach relies on identification of signs and symptoms that correctly classify patients suspected of having the disorder. Truncal obesity, ecchymoses, plethora, proximal muscle weakness, and osteopenia are useful discriminant indices for Cushing’s syndrome, with osteoporosis, ecchymoses, and muscle weakness being the most reliable.69,73,74
FIGURE 15-3. Progression of cushingoid features as shown in photographs taken at 1-year intervals (A through D, progress from earliest to latest).
Increased deposition of fat, one of the earliest signs, occurs in almost all patients and is reported as increasing weight or difficulty in maintaining weight. The distribution of fat is altered in both men and women, with increased amounts in the visceral compartments75 and subcutaneous sites on the face and neck. Increased intra-abdominal fat results in the truncal obesity described by Cushing in approximately 50% of patients. Increased fat in the face (moon facies), the supraclavicular or temporal fossae, and the dorsocervical area (“buffalo hump”) is uncommon in normal people. When extreme, the supraclavicular fat may present as a “collar” rising above the clavicles (Fig. 15-4); filling of the temporal fossae may prevent eyeglass frames from seating properly. Abnormal fat deposition may occur in the epidural space. Spinal epidural lipomatosis causing neurologic deficit, a rare complication of long-term exogenous steroid use, has been reported in a few patients with endogenous Cushing’s syndrome.76,77 Lumbosacral findings were seen in both men and women, whereas thoracic obstruction was restricted to men. The condition can be diagnosed by magnetic resonance imaging (MRI).78
FIGURE 15-4. Fat may fill or, in this case, rise above the supraclavicular fossa of patients with Cushing’s syndrome.
Loss of subcutaneous tissue results in a variety of skin abnormalities that are unusual in the general population and suggest hypercortisolism. Ecchymoses, often after minimal trauma, and cutaneous atrophy, seen as a fine “cigarette paper” wrinkling or tenting over the dorsum of the hand and elbows, are typical. Cutaneous atrophy is influenced by gender and age, with men and the young having greater skin thickness. Two maxims follow: First, it is useful to compare the patient’s skin with that of a near age- and gender-matched healthy person; and second, skin thickness is relatively preserved in cushingoid women with increased androgen production or preservation of ovarian function (Fig. 15-5).
FIGURE 15-5. Thinning of the skin may be demonstrated by twisting the skin on the dorsum of the hand.
Facial plethora, especially over the cheeks, also reflects loss of subcutaneous tissue. Although plethora is more obvious in pale Caucasian individuals, it may be present and should be sought in darker-skinned persons. Because erythema may be induced in normal persons by ultraviolet radiation from lamps or sunlight, wind, or medications (including topical drying agents, glucocorticoids, and antipsoriatic treatments), exposure to these agents should be ascertained before plethora is ascribed to endogenous hypercortisolism. A demarcation line, representing collar, sleeve, or shoulder straps, may differentiate exogenous from endogenous causes. Flushing caused by other conditions (e.g., mastocytosis, thyrotoxicosis, vasomotor instability or estrogen insufficiency in women, carcinoid syndrome) also should be considered.
Purple striae more than 1 cm in diameter are virtually pathognomonic for Cushing’s syndrome (Fig. 15-6). Although the silvery, healed striae that are typical postpartum are not caused by active Cushing’s syndrome, other pink, less pigmented, and thinner striae may be seen. Although most common over the abdomen, striae occur also over the hips, buttocks, thighs, breasts, and upper arms. The tear in the subcutaneous tissue may be best appreciated by indirect (side) lighting, which throws the striae into relief, or by light stroking of the skin. The violaceous hue is not dependent on ACTH-dependent pigmentation and may be seen in Cushing’s syndrome in association with primary adrenal causes.
FIGURE 15-6. Typical abdominal striae of a patient with hypercortisolism. These are greater than 1 cm in width and are violaceous.
Proximal muscle weakness with preservation of distal strength is a hallmark of Cushing’s syndrome. Histologically, this is reflected in profound atrophy of fibers without necrosis.79–81 Weakness is best assessed historically by questions related to the use of these muscles: Is there difficulty or weakness in climbing stairs, getting up from a chair or bed without using hand propulsion, or performing activities using the shoulders (e.g., brushing hair, reaching objects in overhead cabinets, changing ceiling light bulbs)? Formal muscle testing is useful. Assess the strength of the hip flexors by asking the patient to get out of a chair without using his or her arms. If this can be done, the patient is asked to rise from a squat. Inability to perform either task, in the absence of hip or lower extremity arthropathy or other myopathic processes, is suggestive of Cushing’s syndrome. Leg extension while seated is a quantifiable test of proximal muscle strength. The number of seconds for which this position is held can be used to judge deterioration or progress after treatment.
Osteopenia is common. A history of fractures, typically of the feet, ribs, or vertebrae, may be one of the only signs of Cushing’s syndrome, especially in men.70,71,82 Avascular necrosis of bone, a rare complication of endogenous hypercortisolism, is more common in iatrogenic hypercortisolism.83,84 It usually occurs in the hips, but we have also seen it in the knees.
Vellous hypertrichosis of the forehead or upper cheeks distinguishes Cushing’s syndrome from the more common causes of hirsutism and may be appreciated only by careful visual and tactile inspection (Fig. 15-7). Excessive terminal hair on the face and body, and acne—pustular, reflecting increased androgens, or papular, reflecting pure glucocorticoid excess—may be present.85 Severe hirsutism and virilization are uncommon and suggest adrenal carcinoma.
FIGURE 15-7. Vellous hirsutism, especially on the cheeks, is often present in women with Cushing’s syndrome.
Most patients experience emotional and cognitive changes (including increased fatigue, irritability, crying, and restlessness, depressed mood, decreased libido, insomnia, anxiety, impaired memory, concentration, and verbal communication) and changes in appetite. These changes correlate with the degree of hypercortisolism.86 Irritability, characterized as a decreased threshold for uncontrollable verbal outbursts, may be one of the earliest symptoms. Global impairment in neuropsychological function correlates well with the performance of seven serial subtractions and recall of the names of three cities—bedside tests that can be used by the clinician to quantify this symptom complex.87 Approximately 80% of patients meet strict criteria for a major affective disorder—50% with unipolar depression and 30% with bipolar illness.88,89 Although the quality of the depressed mood ranges from suicide attempts to sadness, the time course is characteristically intermittent, rarely lasting longer than 3 days, in contrast to the constant dysphoria reported by depressed patients without Cushing’s syndrome.86 A minority of patients are manic. The improvement in neuropsychiatric findings after treatment of Cushing’s syndrome, coupled with similar features in patients treated with exogenous steroids, and the association of hypercortisolism with poor cognitive performance in depressed patients suggest glucocorticoid excess as a cause.90,91
Hypertension is present in approximately 80% of patients, and although hypertension is also common in the general population, its presence in patients younger than 40 years of age, especially if difficult to control, may alert one to the syndrome. Hypertension usually resolves with treatment of the Cushing’s syndrome but may persist, possibly as the result of microvessel remodeling and/or underlying essential hypertension.92
The association of hypercortisolism and fungal infections of the skin, such as mucocutaneous candidiasis and pityriasis versicolor, with poor wound healing is a common feature. Wound dehiscence occurs less often but is an important consideration in patients who are treated surgically without medical pretreatment.
Patients with marked hypercortisolism (plasma cortisol >43 µg/dL [1200 nmol/L], UFC >2000 µg/day [5520 nmol/day]) are at risk for two potentially catastrophic events: perforation of the viscera and severe infection, either bacterial or opportunistic, such as Pneumocystis carinii, aspergillosis, nocardiosis, cryptococcosis, histoplasmosis, and Candida.93–95 Classic clinical signs, such as loss of bowel sounds and fever, may be absent in peritonitis, and the typical leukocytosis of hypercortisolism may not increase further. Thus, the threshold of suspicion for opportunistic infection and a surgical abdomen must be low in patients with severe hypercortisolism.
Libido is decreased uniformly in men and to a lesser extent (44%) in women,70 in whom increased libido may indicate excess androgen production by an adrenocortical carcinoma. Menstrual irregularities, amenorrhea, and infertility are common and may be the presenting complaints.96 Impotence is common.
PATHOLOGY
The cardinal laboratory findings in endogenous Cushing’s syndrome reflect overproduction of glucocorticoids. Although morning plasma cortisol values may be normal, an increased nighttime nadir blunts or obliterates the normal diurnal rhythm.97–99 This increase in mean 24-hour plasma values is reflected in increased levels of free, or unbound, cortisol in urine100 and saliva.101 The capacity of corticosteroid-binding globulin for cortisol is exceeded at a serum cortisol value of approximately 20 µg/dL (≈600 nmol/L). At this point, the excretion of free cortisol increases dramatically in direct proportion to the increased unbound circulating cortisol values.
Hypokalemic metabolic alkalosis usually is observed when daily urine cortisol excretion is greater than 1500 µg (4100 nmol), and thus mainly in cases of ectopic ACTH syndrome.102 This probably represents a mineralocorticoid action of cortisol at the renal tubule due to saturation of the enzyme 11β-hydroxysteroid dehydrogenase type 2, which inactivates cortisol to cortisone.103 However, although a common feature of ectopic ACTH secretion, it also may occur in approximately 10% of patients with Cushing’s disease. Serum albumin is inversely correlated with cortisol levels, but this is of clinical significance only at very high cortisol levels, and it reverses with treatment for Cushing’s syndrome.104 Drastic reductions in serum albumin should alert the physician to the possibility of concomitant pathology such as infection. Circulating elevated glucocorticoids increase clotting factors, including factor VIII, fibrinogen, and von Willebrand factor, and reduce fibrinolytic activity, resulting in a fourfold risk of thrombotic events.105–107 Lipid abnormalities show increases in very–low density lipoprotein, low-density lipoprotein, high-density lipoprotein, and consequently total cholesterol and triglycerides. These changes probably are caused by a direct cortisol effect of increased hepatic synthesis of very low–density lipoprotein without altered clearance.108,109
Cushing’s syndrome is characterized by insulin resistance and hyperinsulinemia, with frank diabetes mellitus occurring in 30% to 40% of patients, and glucose intolerance in a further 20% to 30%.110,111 A recent study has suggested that as many as 2% of overweight, poorly controlled patients with diabetes may have occult Cushing’s syndrome if fully investigated.112 In the absence of clinical suspicion, the yield is probably lower.113
Patients with Cushing’s disease show accelerated cardiovascular disease, including increased carotid artery intima-media thickness and atherosclerotic plaques on Doppler ultrasonography.114 This increased risk is maintained even as long as 5 years after cure of the hypercortisolemia is attained.115 It also is likely that glucocorticoids have a direct pathogenic effect on the myocardium.
Hypercortisolism suppresses the thyroidal, gonadal, and growth hormone axes. Thyrotropin-releasing hormone and thyroid-stimulating hormone release is disturbed, and particularly the nocturnal surge of thyroid-stimulating hormone is lost, resulting in reduced total thyroxine, total triiodothyronine, and free triiodothyronine levels compared with controls.116 Others have found no differences in free thyroxine or free triiodothyronine levels but have shown a significantly increased prevalence of autoimmune thyroid disease in patients treated for Cushing’s syndrome.117,118 In both men and women, low levels of luteinizing hormone, follicle-stimulating hormone, and gonadal steroids consistent with hypogonadotropic hypogonadism are common and correlate with the degree of hypercortisolemia.119,120 In addition, the coexistence of polycystic ovarian syndrome in Cushing’s syndrome may be more common than was previously thought.96 Hypercortisolemia causes reduced growth hormone (GH) secretion during sleep and blunted GH response to stimulation tests.121
The prevalence of osteoporosis as assessed by dual-energy x-ray absorptiometry is approximately 50% in adult Cushing’s syndrome.122 It appears more common in adrenal Cushing’s syndrome than in Cushing’s disease, and this may relate to the protective effect of increased adrenal androgens in the latter.123
The accentuated visceral fat distribution characteristic of Cushing’s syndrome can be marked when visualized by computed tomography (CT),75 and the liver frequently (20%) is steatotic on imaging.124
CLINICAL SPECTRUM
The typical patient with Cushing’s disease presents at midlife complaining of the gradual development of symptoms, although males tend to present at an earlier age and with more severe clinical consequences.125 Hypokalemia, virilization, and extremely high cortisol excretion (>10-fold normal) are distinctly uncommon and should alert the physician to an alternative cause. The clinical presentation of pituitary corticotroph macroadenomas, apart from visual field changes caused by suprasellar expansion, is not unique. By contrast, invasive pituitary adenomas present at a slightly younger age; cavernous sinus and dural involvement may result in cranial neuropathies and facial neuralgia.126,127 Only a few case reports attest to cerebrospinal or extracranial metastasis of ACTH-producing pituitary tumors.128
Nelson’s syndrome is characterized by the development of hyperpigmentation and high ACTH levels after bilateral adrenalectomy for Cushing’s disease. Tumor growth after adrenalectomy has been attributed to the relative resistance of these tumors to physiologic glucocorticoid suppression.
An abrupt onset of severe Cushing’s syndrome should prompt an evaluation for ectopic ACTH secretion. This variant of ectopic ACTH secretion classically presents as a paraneoplastic syndrome in the context of a known malignancy. The features were captured in the initial formulation of Liddle4: weight loss, hypokalemia, weakness, and diabetes. However, Cushing’s syndrome caused by less obvious ectopic ACTH secretion often presents in the more classic way with weight gain and striae and can be difficult to differentiate clinically from Cushing’s disease. It is patients with this syndrome who most often present a diagnostic dilemma. They tend to have UFC excretion in the range seen in pituitary disease and may not show hypokalemia, hyperpigmentation, or the other findings typical of severe classical ectopic ACTH secretion.
Adrenocortical carcinomas are inefficient producers of cortisol and tend to evince Cushing’s syndrome when the tumor is large (>6 cm), if at all. Abdominal pain or a palpable mass suggests this cause. Feminization in a man or virilization and increased libido in a woman, indicating involvement of the zona reticularis, suggest adrenal cancer or macronodular adrenal disease, which is rarer. The typical patient with PPNAD is a child or young adult who may present with an intermittent course or a family history of associated signs: Lentigines may be the initial clue to this cause. By contrast, patients with the massive macronodular variant of ACTH-independent Cushing’s syndrome tend to be older than 40 years.
DIAGNOSIS AND DIFFERENTIAL DIAGNOSIS
The diagnosis of Cushing’s syndrome rests on the demonstration of both physical and biochemical features of glucocorticoid excess. Thus, the diagnosis is unequivocal in a typical patient, with many of the physical features discussed earlier in the setting of UFC levels more than fourfold above normal.129 However, many of the signs of hypercortisolism, such as obesity, hypertension, glucose intolerance, mood changes, menstrual irregularity, and hirsutism, are common in the general population. Similarly, mild glucocorticoid excess is seen in affective disorders,130 strenuous exercise,59 alcoholism and alcohol withdrawal states,131 renal failure,132 and hypoglycemia. Diagnostic strategies for distinguishing between these pseudo-Cushing’s states and true Cushing’s syndrome are discussed later.
Glucocorticoid resistance is characterized by an abnormal glucocorticoid receptor number or binding, which causes compensatory increases in ACTH and excessive glucocorticoid production to maintain normal glucocorticoid-mediated effects at the target tissues. The diagnosis should be considered in the hypokalemic, hypertensive, hypercortisolemic patient without typical glucocorticoid-mediated signs of Cushing’s syndrome.133
ESTABLISHING THE DIAGNOSIS OF CUSHING’S SYNDROME
When a careful history and physical examination reveal clinical features that could be consistent with the syndrome, exogenous glucocorticoid use must be excluded (Table 15-4). In addition to inquiring about the use of oral, rectal, inhaled, injected, or topical glucocorticoid administration, it is important to evaluate the use of “tonics,” herbs, and skin bleaching creams, which may contain glucocorticoids. In the absence of exogenous glucocorticoids, biochemical confirmation of the diagnosis of Cushing’s syndrome is needed. It is important to remember that the urgency for diagnosis and treatment of Cushing’s syndrome is greatest when the symptoms are severe. In milder cases, the patient may be best served by waiting until the diagnosis is clear. Periodic reevaluation with urine screening tests and documentation of body habitus with photographs may reveal progression.
Table 15-4. Evaluation of Suspected Cushing’s Syndrome
CRH, Corticotropin-releasing hormone; LDDST, low-dose dexamethasone suppression test.
Initial Screening Tests
Hypercortisolemia, demonstrated by loss of the normal circadian rhythm of cortisol secretion, and disturbed feedback of the hypothalamic-pituitary-adrenal (HPA) axis are the cardinal biochemical features of Cushing’s syndrome. Tests to confirm the diagnosis are based on these principles. To screen for Cushing’s syndrome, tests of high sensitivity should be used initially to avoid missing milder cases. All of these screening tests may miss identification of mild cases of hypercortisolemia, and multiple samples or a combination of tests may be needed. A recent guideline suggests that two abnormal first-line test results should be required for the diagnosis of Cushing’s syndrome.134
Urinary Free Cortisol
Under normal conditions, 10% of plasma cortisol is free or unbound and physiologically active. Unbound cortisol is filtered by the kidney, with most being reabsorbed in the tubules and the remainder excreted unchanged. Thus, 24-hour UFC collection produces an integrated measure of serum cortisol, smoothing out variations in cortisol during the day. UFC determinations first became clinically available in 1968135 and have superseded the historical measurement of urinary metabolites of glucocorticoids and androgens (17-hydroxycorticosteroids [17-OHCS], 17-ketosteroids, and 17-ketogenic steroids).
The major drawback of the test is the potential for overcollection or undercollection of the 24-hour specimen, and written instructions must be given to the patient. In addition, creatinine excretion in the collection can be measured to assess completeness and should equal approximately 1 g per 24 hours in a 70 kg patient (variations depend on muscle mass). This value should not vary by more than 10% between collections in the same individual.136 It cannot be used to correct for incomplete collection, however, because rates of cortisol and creatinine excretion are not parallel over the 24-hour period. Various groups have tried to overcome the collection issue by proposing shorter collection periods, usually at night, when the loss of circadian rhythm differs most from normal controls,137,138 but this approach has not been widely accepted. High-performance liquid chromatography and tandem mass spectrometry are now used to measure UFC, which overcomes the previous problem of cross-reactivity of some exogenous glucocorticoids and other structurally similar steroids with conventional radioimmunoassay.139 Occasionally, substances such as carbamazepine, digoxin, and fenofibrate can coelute with cortisol during high-performance liquid chromatography, causing falsely elevated results.140,141
If the previous caveats have been satisfied, the UFC measurement can be interpreted. In large series, measurement of an elevated UFC above the normal range has a high sensitivity for the diagnosis of Cushing’s syndrome (≈95% to 100%).100,142 However, it should be noted that in the latter study, 11% of 146 patients with proven Cushing’s syndrome had at least one of four UFC collections within the normal range, which confirms the need for multiple collections. Values greater than fourfold normal are rare except in Cushing’s syndrome. Values between this and down to the upper limit of normal are compatible with Cushing’s syndrome or pseudo-Cushing’s states, so that one must exclude the latter diagnosis. In summary, UFC measurements have a high sensitivity if collected correctly, and several completely normal collections make the diagnosis of Cushing’s syndrome very unlikely. However, when biochemical evidence of Cushing’s syndrome is not obtained in the setting of clinical features that suggest the diagnosis, repeated measurement of urine cortisol may demonstrate cyclicity or progression. The specificity is somewhat lower, thus patients with marginally elevated levels require further investigation.55
Late-Night Salivary Cortisol
Salivary cortisol measurement offers an excellent reflection of the plasma free cortisol concentration in health and disease because it circumvents the changes in total cortisol due to corticosteroid-binding globulin alterations.143,144 Salivary cortisol is stable for some days at room temperature, and the simple noninvasive collection procedure means that it can be performed conveniently at home and delivered via mail. Thus, it offers a number of attractive advantages over blood collection, particularly in children. Analysis is performed using a modification of the plasma cortisol radioimmunoassay, enzyme-linked immunosorbent assay, or liquid chromatography/tandem mass spectrometry, and commercial kits are internationally available for this.145 The diagnostic value cutoff varies between studies (0.13 µg/dL [3.6 nmol/L] to 0.55 µg/dL [15.2 nmol/L]) because of different assays and comparison groups studied.146–153 Normal values also differ between adult and pediatric populations, and this may be affected by other comorbidities such as diabetes and hypertension.154 However, from these studies, the sensitivity and specificity of this test appear to be relatively consistent at different centers, ranging from 92% to 100%, and from 93% to 100% respectively. It does not appear to make a difference if sampling is done at bedtime (≈23.00 hr) or at midnight, although it should be determined that the patient has a normal sleep pattern. Positive or negative results should be confirmed by repeat sampling. In summary, therefore, although late-night salivary cortisol appears to be a useful and convenient additional screening test for Cushing’s syndrome, particularly in the outpatient setting, local normal ranges should be validated based on the assay used and the population studied.
Low-Dose Dexamethasone Suppression Tests
In normal individuals, administration of the potent synthetic glucocorticoid dexamethasone results in suppression of the HPA axis, whereas patients with Cushing’s syndrome are resistant, at least partially, to this negative feedback. The original low-dose dexamethasone test (LDDST), as described by Liddle in 1960, measured urinary 17-OHCS before and during 48 hours of 0.5 mg dexamethasone every 6 hours, and an excretion of greater than 4 mg/day on the second day of dexamethasone treatment was considered to indicate Cushing’s syndrome.155 Dexamethasone does not cross-react with modern cortisol immunoassays, and the simpler measurement of a single plasma cortisol post dexamethasone has been validated in various series and gives the test a sensitivity of between 97% and 100% for the diagnosis of Cushing’s syndrome.156–159 The simpler overnight LDDST was proposed by Nugent and colleagues in 1965; this measured a 9:00 a.m. plasma cortisol after a single dose of 1 mg dexamethasone taken at midnight.160 Since then, various other doses, between 0.5 and 2 mg, have been proposed for the overnight test, and various diagnostic cutoffs have been applied.161–163 There appears to be no difference in discrimination between single doses of 1, 1.5, and 2 mg.164 Higher doses significantly decrease the sensitivity of the test.165 In a comprehensive review of the LDDST, both the original 2 day test and the 1 mg overnight protocol appear to have comparably high sensitivities (98% to 100%), provided a conservative postdexamethasone serum cortisol cutoff of 1.8 µg/dL (50 nmol/L) is applied. However, the specificity of the overnight test (88%) is lower compared with the 2 day test, particularly if serum cortisol is measured at both 24 and 48 hours (97% to 100%), with potential misclassification of patients with pseudo-Cushing’s states and acute or chronic illnesses. Many endocrinologists use the overnight test because of its greater simplicity and lower cost, although some centers still advocate the 48-hour test because of its high sensitivity and specificity, and the information it can provide in the differential diagnosis of ACTH-dependent Cushing’s syndrome (see later).159 Written instructions should be given to the patient if the latter is to be performed on an outpatient basis. Salivary rather than serum cortisol has been evaluated as the end point for the LDDST. This offers potential benefit in terms of convenience but requires further evaluation.149,166
Factors such as variable absorption and increased or decreased dexamethasone metabolism due to other compounds (Table 15-5) can influence any oral dexamethasone test.167 Therefore, a history of symptoms of malabsorption and a careful drug history should be taken before the test is used in a patient. Measurement of plasma dexamethasone is available in some centers and can be useful in patients of concern. One solution to overcome demonstrated malabsorption is to use one of the published intravenous dexamethasone suppression tests, recognizing that criteria for response have not been standardized.168,169 Pregnancy and other causes of increased or decreased corticosteroid-binding globulin (such as exogenous estrogens and the nephrotic syndrome) also should be excluded because these are likely to result in false-positive and false-negative tests.170
Table 15-5. Spurious Causes of Abnormal Dexamethasone Suppression Test Results
| False Positive |
Increased metabolism: barbiturates, phenytoin, carbamazepine, primidone, rifampicin, aminoglutethimide |
| False Negative |
Second-Line Tests
The Dexamethasone-CRH Test
In 1993, a combined dexamethasone-CRH (Dex-CRH) test was introduced for the difficult scenario of the differentiation of pseudo-Cushing’s states from true Cushing’s syndrome in patients with only mild hypercortisolemia and equivocal physical findings.171 Dexamethasone 0.5 mg every 6 hours was given for eight doses, ending 2 hours before administration of ovine CRH (1 µg/kg intravenously) to 58 adults with UFC less than 360 µg/day (<1000 nmol/day). Subsequent evaluation proved that 39 had Cushing’s syndrome and 19 had a pseudo-Cushing’s state. The plasma cortisol value 15 minutes after CRH was less than 1.4 µg/dL (38 nmol/L) in all patients with pseudo-Cushing’s states and was greater in all patients with Cushing’s syndrome. A prospective follow-up study by the same group in 98 patients continued to show that the test had an impressive sensitivity and specificity of 99% and 96%, respectively.172 However, results from a number of other smaller studies have challenged the diagnostic utility of this test over the standard LDDST.173–175 Overall, in these reports, the specificity of the LDDST in 92 patients without Cushing’s syndrome was 79%, versus 70% for the Dex-CRH. Test sensitivity in 59 patients with Cushing’s syndrome was 96% for LDDST, versus 98% for the Dex-CRH group. It perhaps is not surprising that the diagnostic utility of the Dex-CRH has altered with additional studies at a greater number of centers. This might be the case for a number of reasons, including variable dexamethasone metabolism in individuals, different definitions of patients with pseudo-Cushing’s, different protocols and assays, and variable diagnostic thresholds.176 Of note, the original cortisol criteria performed poorly at these other centers, and this may have happened because many cortisol assays do not reliably measure levels <1.8 µg/dL (50 nmol/L). It does highlight that as a clinician one must be confident in the assay that is to be used for a particular test, and diagnostic criteria should be chosen that are appropriate for that assay. The Dex-CRH test remains a test that should be considered in patients with equivocal results.
Plasma Cortisol Circadian Rhythm
The normal diurnal rhythm of plasma cortisol is blunted or absent in Cushing’s syndrome, with normal or increased morning values and an increase in the nighttime nadir. Although less convenient than salivary cortisol, midnight plasma cortisol levels may be useful to obtain in patients admitted for investigation. Samples are best obtained around midnight, through an indwelling line for awake patients or by direct venipuncture within 5 to 10 minutes of waking of sleeping patients. In one study, 20 normal sleeping subjects had values less than 1.8 µg/dL (50 nmol/L), whereas all 150 patients with Cushing’s syndrome had midnight plasma cortisol concentrations greater than this.158 The suggested cutoff criterion in awake patients is higher, 7.5 to 8.3 µg/dL (207 to 229 nmol/L) and less discriminatory (sensitivity 92% to 94%, and specificity 96% to 100%).177,178 This difference probably reflects a different comparison group—patients suspected to have Cushing’s but in whom it was excluded. Patients with severe medical illness, depression, and mania may have cortisol values one to three times normal.130,164 Therefore, a sleeping midnight cortisol value less than 1.8 µg/dL (50 nmol/L) effectively excludes active Cushing’s syndrome, but higher values, unless very high, are less specific for Cushing’s syndrome.
Other Second-Line Tests
The insulin tolerance test has been used to distinguish Cushing’s syndrome from pseudo-Cushing’s states. Serum cortisol values increase in normal people after acute hypoglycemia, presumably because of central stimulation of CRH and vasopressin. The sustained hypercortisolism of Cushing’s syndrome suppresses CRH and vasopressin secretion and so blunts this response. The CRH/vasopressin neurons are presumed to be overactive in pseudo-Cushing’s states, particularly those that are depression associated, so a normal response to hypoglycemia (<40 mg/dL; <2.2 nmol/L) usually is maintained. Unfortunately, approximately 18% of patients with Cushing’s syndrome, especially those with minimal hypercortisolism, show a normal response to adequate hypoglycemia.164 Additionally, criteria for interpretation of results have not been established. If used, a dose of insulin of 0.3 U/kg should be used to overcome insulin resistance in these patients.130
The opiate agonist loperamide (16 mg orally) has been shown to inhibit CRH and thus ACTH and cortisol levels in most normal individuals, but not in patients with Cushing’s syndrome. This test has not been used widely but has been evaluated in one center, revealing a sensitivity of 100% and a specificity of 95%.179,180 However, it is unclear as to how well this test may exclude pseudo-Cushing’s states because a significant proportion of patients with depression also fail to suppress the HPA axis.181 It does not appear to be affected by drugs that affect the metabolism of dexamethasone and could potentially be useful in assessing patients on such treatment.180
DIFFERENTIAL DIAGNOSIS OF CUSHING’S SYNDROME
Once the diagnosis of Cushing’s syndrome is made, its cause must be determined. The strategy for the differential diagnosis of Cushing’s syndrome (Fig. 15-8) begins with measurement of plasma ACTH to distinguish between ACTH-dependent and ACTH-independent causes. Modern two-site immunoradiometric assays are more sensitive than the older radioimmunoassays and therefore provide the best discrimination. Only assays that can reliably detect values to below 10 ng/L should be used, and appropriate collection and processing of the sample are essential, because ACTH is susceptible to degradation by peptidases; therefore the sample must be kept in an ice water bath and centrifuged, aliquoted, and frozen within a few hours to avoid a spuriously low result. Repeated measurements are usually necessary because patients with ACTH-dependent Cushing’s disease have been shown to have on occasion ACTH levels less than 10 ng/L (2 pmol/L) on conventional radioimmunoassay,182 but consistent ACTH measurements of less than 10 ng/L (2 pmol/L) at 9:00 am with concomitant hypercortisolemia essentially confirm ACTH-independent Cushing’s syndrome. When the basal ACTH level is indeterminate (10 to 20 g/L [2 to 4 pmol/L]), the response to CRH may be useful in this setting. Patients with primary adrenal disease rarely show maximal ACTH values greater than 20 ng/L (4 pmol/L), although patients with Cushing’s disease usually exceed this value.
FIGURE 15-8. Suggested strategy for the differential diagnosis of Cushing’s syndrome. ACTH, Adrenocorticotropic hormone; AIMAH, ACTH-independent bilateral macronodular adrenal hyperplasia; BIPSS, bilateral inferior petrosal sinus sampling; CRH, corticotropin-releasing hormone; CT, computed tomography; HDDST, high-dose dexamethasone suppression test; MRI, magnetic resonance imaging; PPNAD, primary pigmented nodular adrenal disease.
Investigating Adrenocorticotropic Hormone–Independent Cushing’s Syndrome
Radiologic tests are the mainstay in differentiating between the various types of ACTH-independent Cushing’s syndrome. High-resolution CT scanning of the adrenal glands has excellent diagnostic accuracy for masses greater than 1 cm and allows evaluation of the contralateral gland.183 MRI may be useful for the differential diagnosis of adrenal masses; the T2-weighted signal is progressively brighter in normal tissue, adenoma, carcinoma, and finally pheochromocytoma.184 With this approach, adrenal tumors appear as a unilateral mass with an atrophic or less commonly a normal-size contralateral gland.185 If the lesion is greater than 5 cm in diameter, it should be considered to be malignant until proven otherwise, and imaging characteristics should not be relied upon. Very rarely, bilateral adenomas may be present.186 The adrenal glands in PPNAD appear normal or slightly lumpy from multiple small nodules but generally are not enlarged.187 AIMAH is characterized by bilaterally huge (>5 cm) nodular or hyperplastic glands.188 Exogenous administration of glucocorticoids results in adrenal atrophy; very small glands may provide a clue as to this entity.
The CT appearance of the adrenals in AIMAH may be similar to that of ACTH-dependent forms of Cushing’s syndrome, in which adrenal enlargement is present in 70% of cases.189 However, in our experience, the adrenal glands in Cushing’s disease are smaller and usually are symmetrically enlarged with an occasional nodule, as opposed to large or huge glands with definite nodules in AIMAH. In addition, the two can usually be differentiated by the ACTH level, although some patients with the macronodular subset of Cushing’s disease can develop a degree of adrenal autonomy that can cause biochemical confusion.190 Occasionally, confusion also may arise with apparent unilateral adrenal lesions, when the biochemistry is consistent with an ACTH-dependent cause; we generally would rely on the biochemistry in this situation and would examine the contralateral gland to see whether it is hyperplastic.
Differentiating Between Adrenocorticotropic Hormone–Dependent Causes of Cushing’s Syndrome
Although some patients with ectopic ACTH secretion, usually those with overt tumors, have extremely elevated values of plasma ACTH (>100 ng/L [>20 pmol/L]), complete overlap is seen between values in occult ectopic ACTH secretion and in Cushing’s disease.191 Therefore, ACTH values alone cannot differentiate reliably the ACTH-dependent forms of Cushing’s syndrome.
The ACTH-dependent forms of Cushing’s syndrome present the greatest diagnostic challenge. Cushing’s disease accounts for by far the majority of cases of ACTH-dependent Cushing’s syndrome—overall approximately 80% to 90% in most series. This percentage is gender dependent and is higher in women than in men,192 although in prepubertal childhood, an anomalous 80% male preponderance is noted. Therefore, even before one starts further investigation, the pretest probability that the patient has Cushing’s disease is very high, and any investigation must improve on this. The specificity of any test should be as close to 100% as possible for the diagnosis of Cushing’s disease, to avoid inappropriate pituitary surgery in patients with ectopic ACTH production. A variety of functional tests of the HPA axis have been developed to take advantage of the differences in pathophysiology between ACTH-dependent causes of Cushing’s syndrome. Some of these investigations have evolved, and others have fallen by the wayside.
Bilateral Inferior Petrosal Sinus Sampling
Bilateral inferior petrosal sinus sampling (BIPSS) is the best test for distinguishing ACTH-dependent forms of Cushing’s syndrome, as long as the patient has active hypercortisolemia, which should be confirmed at the time of the procedure.193,194 The test exploits the normal venous drainage of each half of the pituitary gland via the cavernous sinus into the corresponding petrosal sinus. Each petrosal sinus is catheterized separately via a femoral approach, and blood for measurement of ACTH is obtained simultaneously from each sinus and a peripheral vein at two timepoints before and at 3 to 5 minutes and possibly also 10 minutes after the administration of ovine or human CRH (Ferring) (1 µg/kg or 100 µg intravenously)195 (Fig. 15-9). Where CRH is unavailable for whatever reason, recent data suggest that 10 µg desmopessin may be a suitable alternative.
FIGURE 15-9. Maximal ratio of adrenocorticotropic hormone (ACTH) concentration in the inferior petrosal sinus to peripheral blood in patients with confirmed Cushing’s disease, ectopic ACTH syndrome, or adrenal disease before corticotropin-releasing hormone (CRH) (A) or at any time before or after CRH administration (B). A ratio of 3.0 had 100% sensitivity and specificity.
(Data from Oldfield EH, Doppman JL, Nieman LK, et al: Petrosal sinus sampling with and without corticotropin-releasing hormone for the differential diagnosis of Cushing’s syndrome. N Engl J Med 325:897–905, 1991.)
ACTH concentrations are greater in the central samples in Cushing’s disease and increase after CRH administration, reflecting ACTH secretion by the corticotroph adenoma. In contrast, ACTH values in the central and peripheral specimens are similar in ectopic ACTH secretion and do not increase after CRH. A ratio of central (i.e., petrosal) to peripheral ACTH values is calculated. In earlier series, pre-CRH ratios greater than 2 or post-CRH ratios greater than 3 were 100% specific for Cushing’s disease,196,197 but a small number of false positives have been reported in later series.198,199 The sensitivity of the test is improved after CRH; however, false negatives still occur in about 6% of patients with Cushing’s disease. False positives in ectopic ACTH are extremely rare.
It should be remembered that the technique is highly specialized, and allied with this are a number of important points. First, both petrosal sinuses must be cannulated adequately and catheter placement confirmed before and after sampling.200 Second, the radiologist must confirm the venous anatomy because anomalous venous drainage can give false-negative results. Preliminary data have suggested that simultaneous measurement of prolactin can be used as an index of pituitary venous drainage, and prolactin should be measured when results indicate a noncentral source of ACTH.201 Third, the procedure carries a small risk of complications. Transient ear discomfort or pain can occur, as can local groin hematomas. More serious transient and permanent neurologic sequelae, including brain stem infarction, have been reported, although these are rare (<1%), and most have been related to the particular type of catheter used202,203; if any early warning signs of such events are observed, the procedure should be halted immediately. Patients should be given heparin during sampling to prevent thrombotic events.129 CRH itself generally is tolerated well, although patients may experience brief facial flushing and a metallic taste in the mouth. One case of CRH induced pituitary apoplexy in a patient with Cushing’s disease has been reported.204
Another potential advantage of BIPSS involves lateralizing microadenomas within the pituitary using the inferior petrosal sinus ACTH gradient, with a basal or post-CRH intersinus ratio of at least 1.4 being the criterion used for lateralization in all large studies.196,197,205,206 In these studies, the diagnostic accuracy of localization as assessed by operative outcome varied between 59% and 83%. This is improved if venous drainage is assessed to be symmetric.207 Some discrepancy has been noted between studies as to whether CRH improves the predictive value of the test.208 If a reversal of lateralization is seen pre- and post-CRH, the test cannot be relied upon.209 Sampling of the internal jugular veins is a simpler procedure but is not as sensitive as BIPSS.210 However, it may be a useful technique in less experienced centers, with the caveat that patients with negative results then are referred for BIPSS.211 In our opinion, sampling from the cavernous sinus itself offers no great advantage.
High-Dose Dexamethasone Suppression Test
The original high-dose dexamethasone suppression test (HDDST) was described in the same paper as the 48-hour LDDST; 2 mg dexamethasone is used in place of 0.5 mg, with a 50% reduction in urinary 17-OHCS shown to differentiate 96% of patients with Cushing’s disease rather than adrenal tumors.155 The role of the HDDST in the differential diagnosis of ACTH-dependent Cushing’s syndrome is based on the same premise, that is, that most pituitary corticotroph tumors retain some responsiveness (albeit reduced) to negative glucocorticoid feedback on ACTH secretion, whereas ectopic ACTH-secreting tumors, like adrenal tumors, typically do not. Measurement of UFC or plasma/serum cortisol has superseded that of urinary 17-OHCS, and an overnight test has been advocated, with a single dose of 8 mg dexamethasone given at 11:00 pm, and with the criterion of a 50% reduction in plasma cortisol levels on the morning after administration.212 Despite evidence that only about 80% of patients with Cushing’s disease will show suppression of plasma cortisol to less than 50% of the basal value, and large numbers of patients with ectopic Cushing’s syndrome have false-positive results (≈30%),60,213 the HDDST is still used widely. Some data suggest that that suppression to HDDST can be inferred by a greater than 30% suppression of serum cortisol to the 2 day LDDST (Fig. 15-10); therefore, in centers that use this form of the LDDST, the HDDST may not confer any extra information.159 It should not be forgotten that patients are receiving large doses of glucocorticoids, in addition to their high endogenous cortisol production, and one should be alert for the precipitation of psychosis and/or worsening of glycemic control or other complications.
FIGURE 15-10. Correlation between the degree of suppression during the low-dose dexamethasone suppression test (LDDST) and during the high-dose dexamethasone suppression test (HDDST) in 185 patients with Cushing’s disease.
(Reproduced with permission from Isidori AM, Kaltsas GA, Mohammed S, et al: Discriminatory value of the low-dose dexamethasone suppression test in establishing the diagnosis and differential diagnosis of Cushing’s syndrome. J Clin Endocrinol Metab 88:5299–5306, 2003.
Corticotropin-Releasing Hormone Stimulation Test
The use of CRH stimulation for the differential diagnosis of ACTH-dependent Cushing’s syndrome is based on two assumptions: (1) that corticotropinomas retain responsivity to CRH, whereas noncorticotroph tumors lack CRH receptors and cannot respond to the agent; and (2) that hypercortisolism has been sufficient to inhibit the normal corticotroph response. Indeed, most patients with Cushing’s disease respond to CRH, either 1 µg/kg or 100 µg intravenous synthetic ovine or human sequence CRH, with increases in plasma ACTH or cortisol, and patients with ectopic ACTH secretion typically do not.214–216 Human sequence CRH has qualitatively similar properties to ovine CRH, although it is shorter acting with a slightly smaller increase in plasma cortisol and ACTH in normal and obese patients and in those with Cushing’s disease217; this may be related to the more rapid clearance of the human sequence by endogenous CRH-binding protein.218 Availability differs worldwide, with ovine CRH predominant in North America but human CRH elsewhere.
Because different centers have used differing protocols, including different types of CRH and different sampling timepoints, little consensus has been reached on a universal criterion for interpreting the test. However, where the test has been validated in experienced centers, the diagnostic utility appears similar. For instance, in the largest published series of the use of ovine CRH in ACTH-dependent Cushing’s syndrome, an increase in ACTH of at least 35% from a mean basal (5 minutes and 1 minute) to a mean of 15 and 30 minutes after ovine CRH in 100 patients with Cushing’s disease and in 16 patients with ectopic ACTH syndrome (Fig. 15-11) gave the test a sensitivity of 93% for diagnosing Cushing’s disease and 100% specificity. The best cortisol criterion was an increase of at least 20% at a mean of 30 and 45 minutes, revealing a sensitivity of 91% and a specificity of 88%.219 Similarly, in the largest series involving use of the human CRH test in 101 patients with Cushing’s disease and in 14 with ectopic ACTH syndrome, the best criterion used to differentiate Cushing’s disease from ectopic ACTH syndrome was an increase in cortisol of at least 14% from a mean basal (15 and 0 minutes) to a mean of 15 and 30 minutes, yielding a sensitivity of 85% and a specificity of 100% (Fig. 15-12). In contrast, the best ACTH response was a maximal increase of at least 105%, indicating 70% sensitivity and 100% specificity.192 The CRH test is a useful discriminator between causes of ACTH-dependent Cushing’s syndrome; however, which cutoff to use must be evaluated at individual centers, and caution should be exercised because undoubtedly there will be patients with ectopic ACTH syndrome who respond outside these cutoffs. However, an increase in cortisol outside the normal range can differentiate Cushing’s disease from normality, albeit in only approximately 50% of cases, and, as noted previously, the measurement of plasma ACTH in the test can help to discriminate ACTH-dependent from ACTH-independent causes of Cushing’s syndrome when basal levels of ACTH are equivocal. We therefore believe that this test plays a useful role in the investigation of patients with Cushing’s syndrome.
FIGURE 15-11. Response of adrenocorticotropic hormone (ACTH) and cortisol to ovine corticotropin-releasing hormone in patients with Cushing’s disease and ectopic ACTH secretion. ACTH responses are expressed as the percentage of change in mean concentration 15 and 30 minutes after ovine corticotropin-releasing hormone from the mean basal value 1 and 5 minutes before the injection. The dashed line indicates a response of 35%, representing a diagnostic criterion with 100% specificity and 93% sensitivity. Cortisol responses are expressed as percentage of change in mean cortisol concentration 30 and 45 minutes after ovine corticotropin-releasing hormone from the mean basal value 1 and 5 minutes before the injection. The dashed line indicates a response of 20%, representing a diagnostic criterion with 88% specificity and 91% sensitivity.
(Data from Nieman LK, Oldfield EH, Wesley R, et al: A simplified morning ovine corticotropin-releasing hormone stimulation test for the differential diagnosis of ACTH-dependent Cushing syndrome. J Clin Endocrinol Metab 77:1308–1312, 1993.)
FIGURE 15-12. Percentage change in serum cortisol from a mean basal at 15 and 0 minutes to a mean value calculated from the levels at 15 and 30 minutes after the administration of human corticotropin-releasing hormone (100 mg intravenously) in 100 patients with Cushing’s disease (CD) and 14 patients with ectopic adrenocorticotropic hormone syndrome.
(Reproduced with permission from Newll-Price J, Morris DG, Drake WM, et al: Optimal response criteria for the human CRH test in the differential diagnosis of ACTH-dependent Cushing’s syndrome. J Clin Endocrinol Metab 87:1640–1645, 2002.)
Other Stimulation Tests
Vasopressin and desmopressin (a synthetic long-acting vasopressin analogue without V1-mediated pressor effects) are thought to stimulate ACTH release in Cushing’s disease through the corticotroph-specific V3 (or V1b) receptor. Hexarelin (a growth hormone secretagogue) also stimulates ACTH release to a sevenfold greater extent than human CRH; although the mechanism has not been entirely elucidated, this probably occurs through stimulation of vasopressin release in normal subjects,220 and through stimulation of aberrant growth hormone secretagogue receptors in patients with corticotroph tumors.221 These peptides all have been used in a similar manner to CRH, to try to improve the differentiation of ACTH-dependent Cushing’s syndrome, but they generally have proved inferior or of no advantage.222–225 However, desmopressin may be useful in centers where CRH is unavailable. A combined desmopressin (10 µg) and human CRH (100 µg) test initially looked extremely promising.226 However, a later study of this combined test in 26 patients with Cushing’s disease and 5 patients with ectopic ACTH syndrome showed significant overlap in responses.227 The disappointing discriminatory outcome of these stimulants is undoubtedly due to the expression of both vasopressin and growth hormone secretagogue receptors by some ectopic ACTH-secreting tumors.129,228
Measurement of marker peptides, such as calcitonin, gastrin, 5-hydroxyindoleacetic acid, serotonin, and catecholamines or their metabolites, may help to identify a neuroendocrine tumor.
Combined Test Strategies
Because none of the noninvasive tests have 100% diagnostic accuracy, a number of investigators have evaluated the utility of combined test strategies. The CRH and the HDDST have been paired in this way, and combined, they have a diagnostic accuracy greater than that of either test alone, yielding 98% to 100% sensitivity and 88% to 100% specificity.229–231 Similar high accuracy has been obtained by combining the results of the LDDST and the CRH test.159
Imaging of the Adrenocorticotropin Hormone Source
Pituitary MRI imaging before and after gadolinium enhancement should be performed in all patients with ACTH-dependent Cushing’s syndrome via T1-weighted spin echo and/or spoiled gradient recalled acquisition (SPGR) techniques. These will identify an adenoma in up to 80% of patients with Cushing’s disease,60,232,233 and in approximately 10% of normal individuals.234 Most adenomas (95%) exhibit a hypointense signal with no postgadolinium enhancement, and the remaining 5% show an isointense signal post gadolinium enhancement.235 CT imaging typically shows a hypodense lesion that fails to enhance post contrast but is less sensitive than MRI in detecting small (<5 mm) adenomas and is not recommended for this reason.60,236
Imaging is the most helpful way to identify the source of ectopic ACTH production. Given the likely sites of tumors, CT and/or MRI of the neck, chest, and abdomen should be obtained. The most common source is a bronchial carcinoid tumor, but small (<1 cm) lesions often can prove difficult to locate. Fine-cut high-resolution CT scanning with both supine and prone images can help differentiate between tumors and vascular shadows.55 MRI can identify chest lesions that are not evident on CT scanning and that characteristically show a high signal on T2-weighted and short-inversion time inversion recovery (STIR) images.237 Additionally, pheochromocytomas are bright on T2-weighted MRI. CT-guided aspiration of masses for measurement of ACTH may provide useful functional information.238
Because most ectopic ACTH-secreting tumors are of neuroendocrine origin and therefore may express somatostatin receptor subtypes, radiolabeled somatostatin analogue (111In-pentetreotide) scintigraphy may be useful to show functionality of identified tumors, and sporadic reports have indicated that it identifies lesions not apparent on conventional imaging.239–241 However, in most patients, including a recent series of 35 patients with ectopic ACTH secretion, 111In-pentetreotide scintigraphy was not able to detect tumors when the MRI or CT scan was negative, and significant numbers of false-positive scans resulted.242 Thus, CT and MRI represent the best initial screening examinations, but scintigraphy may be a useful adjunctive imaging modality to help confirm abnormalities seen on CT or MRI. 18-Flurodeoxyglucose positron-emission tomography (PET) generally does not offer any advantage over conventional CT or MRI unless the tumors are metabolically active,243 which is not usually the case.244 However, 67Ga-octreotate PET scintigraphy may show advantages over 111In-pentetreotide scintigraphy.245
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
