Two articles on stroke, beginning on page 95
Postgraduate Medicine
Taylor & Francis Taylor & Francis Group
ISSN: 0032-5481 (Print) 1941-9260 (Online) Journal homepage: http://www.tandfonline.com/loi/ipgm20
Cushing’s Syndrome
James V. Felicetta MD
To cite this article: James V. Felicetta MD (1989) Cushing’s Syndrome, Postgraduate Medicine, 86:8, 79-90, DOI: 10.1080/00325481.1989.11704497
To link to this article: http://dx.doi.org/10.1080/00325481.1989.11704497
Published online: 17 May 2016.
Submit your article to this journal
Q
View related articles
Downloaded by [Monash University Library] at 06:28 03 July 2016
Cushing’s syndrome How to pinpoint and treat the underlying cause
James V. Felicetta, MD
Preview
Several different clinical conditions cause Cushing’s syndrome. With few exceptions, all produce the same signs and symptoms. Thus, making a definitive diagnosis can be a time-consuming chore for the already harried primary care physician. In this ar- ticle, Dr Felicetta simplifies the process by describing the most likely underlying causes of glucocorticoid excess, the diagnostic workup, and specific therapeutic approaches.
Cushing’s syndrome results from a marked increase in glucocorti- coids, which can be caused by several different clinical condi- tions. No matter what the under- lying cause, the array of physical findings produced by glucocorti- coid excess is the same. Moreover, with one exception (ie, iatrogenic Cushing’s syndrome), the source of glucocorticoid excess is always the same: the adrenal glands. Given this scenario and the fact that Cushing’s syndrome is less common than some other endo- crine disorders, the busy clinician needs to know when to entertain the diagnosis and how to pinpoint and treat the underlying cause of the condition.
Causes
In addition to prolonged glucocor- ticoid therapy (probably the most common cause), at least five other distinct clinical conditions result in Cushing’s syndrome. The distinguishing factor in each is the specific pathophysiologic process that prompts the adrenal glands to produce glucocorticoids in excess.
EXCESS PITUITARY SECRETION OF CORTICOTROPIN-This patho- logic process causes a subtype of Cushing’s syndrome that is clas- sically known as Cushing’s dis- ease. Roughly 60% to 65% of all patients with noniatrogenic Cushing’s syndrome actually have Cushing’s disease.1 Before the advent of computed tomogra- phy (CT) and magnetic resonance imaging (MRI), only 5% to 10% of all cases of Cushing’s syndrome appeared to be caused by pitu- itary tumors.1 Since then, the percentage of patients who are found to have pituitary microad- enomas has increased.2 However, not all patients with Cushing’s disease have discrete pituitary tumors. Some simply have diffuse
pituitary overproduction of corti- cotropin (ACTH), perhaps driven by a primary hypothalamic dis- ease. Even though glucocorti- coids continue to exert their usual inhibitory effect on the pituitary in these patients, supranormal amounts are required to suppress secretion of corticotropin.
In the days when bilateral adrenalectomy was standard
therapy for Cushing’s disease, corticotropin levels typically remained very high postopera- tively, even when replacement therapy with steroids maintained plasma cortisol at normal levels. This observation confirms the reduced effectiveness of circulat- ing glucocorticoids in suppress- ing pituitary secretion of cortico- tropin.
ECTOPIC PRODUCTION OF CORTI- COTROPIN BY MALIGNANT NONPI- TUITARY TUMORS-Oat cell (small-cell) carcinoma of the lung is responsible for more than one half of cases of this type of Cush- ing’s syndrome; islet cell tumor, thymoma, and bronchial ade- noma account for the remaining cases. Because the disease pro- gresses rapidly, patients with malignant nonpituitary tumors may not exhibit many of the classic characteristics of Cush- ing’s syndrome. Also, the weight loss that typically occurs in pa- tients with malignant neoplasms may mask the usual cushingoid features. Several cases have recently been reported3,4 in which malignant nonpituitary tumors were found to secrete corticotropin-releasing hormone (CRH), a hypothalamic hormone. The CRH secretion in turn stimu- lates pituitary overproduction of corticotropin.
BENIGN ADRENAL ADENOMA In most series, benign adrenal adenomas account for only about 10% of cases of Cushing’s syn- continued
Because of rapid disease progression, many patients with malignant nonpituitary tumors may not exhibit many of the classic characteristics of Cushing’s syndrome.
drome. These tumors usually weigh less than 100 g. They typically synthesize cortisol quite efficiently. Although they are not under pituitary control, many of them do respond to corticotropin stimulation with increased gluco- corticoid output.
ADRENAL CARCINOMA-Ac- cording to Hutter and Kayhoe,5 adrenal carcinoma has usually metastasized by the time the patient initially seeks medical attention. The primary site of metastasis is the liver, but metas- tasis to the lung, gastrointestinal tract, bone, lymph nodes, and mediastinum is also seen. Unlike adrenal adenomas, adrenal carci- nomas synthesize cortisol very inefficiently and often grow very large before cushingoid symp- toms appear. About one half of these tumors are palpable on physical examination. A calcified adrenal mass should raise suspi- cion of a malignant process.
PRIMARY ADRENOCORTICAL NODULAR DYSPLASIA-This new- ly described, somewhat rare cause of Cushing’s syndrome results in the autonomous pro- duction of glucocorticoids. In
James V. Felicetta, MD Dr Felicetta is chief of medicine at the Veterans Affairs Medical Center, Phoe- nix, and associate clinical professor of medicine, University of Arizona College of Medicine, Tucson. His research interests include thyroid hormone metabolism and the role of calcium and other cations in the etiology of hypertension.
their review of the literature, Larsen and colleagues6 identified 30 patients with this condition. All had very low levels of cortico- tropin. Also noted were brown- black nodules up to 5 mm in diameter in the adrenal cortex, with atrophy of the remaining cortex. These distinctive findings clearly differentiate this condition from the bilateral adrenal hyper- plasia seen with excessive corti- cotropin production.
Typical features
As already stated, the features of Cushing’s syndrome are much the same regardless of the under- lying cause.
CLINICAL MANIFESTATIONS Major changes in the overall distribution of body fat are evi- dent. Patients show a definite tendency toward central (or buf- falo) obesity, with the thorax and abdomen showing a much great- er increase in fat than the arms and legs (centripetal obesity). Increased fat distribution in the supraclavicular spaces is also typical, as is the presence of a buffalo hump. Presence of the buffalo hump, however, has only limited clinical usefulness, be- cause it occurs in many diseases other than Cushing’s syndrome.
Also characteristic is a rounded face, the so-called moon facies, with fat above as well as below the zygomatic arches. Chest films sometimes show enlargement of
the mediastinum caused by an increase in perihilar fat. A dewlap on the chest, ie, a fatty skin tag named after a similar-looking structure that is part of the nor- mal anatomy of bovine animals, may also be present.
Almost all patients with Cush- ing’s syndrome have marked muscle wasting. Muscle weak- ness is usually more pronounced proximally than distally.
A loss of elastic fiber gives the skin a thin, fine texture. The face may be red because the trans- parency of thin skin makes the underlying vasculature more visible. Striae over the abdomen, thighs, arms, and chest often occur for the same reason. In addition, patients with Cushing’s syndrome bruise easily, and wound healing is often poor.
Another important feature of Cushing’s syndrome is general- ized osteoporosis, which occurs because glucocorticoids inhibit the activity of osteoblasts. Bone pain and pathologic fractures are common manifestations of Cush- ing’s syndrome. Indeed, the pre- senting symptom in many pa- tients with Cushing’s syndrome is back pain caused by compres- sion fractures of the thoracic and lumbar vertebrae.
Psychiatric disturbances are also common. They range from mild irritability or excitability to frank steroidal psychosis.
If androgen secretion is exces- continued
The presenting symptom in many patients with Cushing’s syndrome is back pain caused by compression fractures of the thoracic and lumbar vertebrae.
sive, female patients may have hirsutism, acne, and amenor- rhea. However, signs of frank virilization, such as clitoromegaly and frontal balding, are unusual. Acne may also develop in male patients.
Hypertension is a common manifestation of Cushing’s syn- drome. In most patients, it ap- pears to be the direct result of glucocorticoid excess. Glucocorti- coids inhibit the synthesis of prostacyclin,7 a potent vasodila- tor. Thus, elevated glucocorticoid levels alter the delicate balance between vasodilating and vaso- constricting influences on the tone of smooth-muscle cells in blood vessels. In some patients, the adrenal cortex secretes ex- cessive amounts of the mineralo- corticoid 11-deoxycorticosterone; hypokalemia and edema result.
LABORATORY FINDINGS-Re- sults of routine blood chemistry studies are usually relatively normal in Cushing’s syndrome. However, the neutrophil count is usually elevated, probably be- cause of demargination of white blood cells caused by glucocorti- coid excess. Lymphocyte and eosinophil counts are relatively low.
Carbohydrate metabolism is typically abnormal in Cushing’s syndrome. Glucocorticoids stim- ulate hepatic glycogenolysis and gluconeogenesis and directly antagonize the actions of insulin
at a number of different sites. Thus, blood glucose control wors- ens in diabetic patients, and carbohydrate intolerance may become apparent in patients who have no history of diabetes.
Diagnosis
In the past, the workup for possi- ble Cushing’s syndrome began with a 24-hour urine collection for measurement of 17-hydroxy- corticosteroid and 17-ketosteroid levels. The 17-hydroxycorticoste- roids primarily reflect glucocorti- coid output, whereas 17-ketoste- roids reflect adrenal androgen output, which may also be in- creased in Cushing’s syndrome. A marked elevation in 17-keto- steroid level is highly suggestive of adrenal carcinoma. Because 17-hydroxycorticosteroid levels are often falsely elevated in obese patients, they are not diagnostic of Cushing’s syndrome. Current- ly, urinary 17-hydroxycorticoste- roid determination has been largely supplanted by 24-hour measurement of urinary free cortisol, which is not elevated by obesity alone.
One of the earliest abnormali- ties seen in Cushing’s syndrome is the loss of the normal diurnal pattern of cortisol secretion. In healthy subjects, corticotropin levels peak during the middle of the night, whereas cortisol levels, which are entrained to corticotro- pin levels, peak early in the
morning. Corticotropin and corti- sol levels both fall progressively throughout the day. The cortisol level at 8 PM is usually less than one third the cortisol level at 8 AM, and the 8-PM level is always less than 10 µg/dL. In contrast, plasma cortisol levels in patients with Cushing’s syndrome do not differ much from morning to evening. Certainly, this finding should raise suspicion of Cush- ing’s syndrome and indicate the need for further workup.
Today, the diagnostic workup for possible Cushing’s syndrome relies heavily on the dexameth- asone (Decadron, Hexadrol) suppression test (figure 1).8.9 Originally, the test was performed using urinary measurement, but plasma measurement is more common today. Dexamethasone testing consists of three separate stages; however, depending on results, all three stages may not be needed.
The first stage is the overnight screening suppression test, in which the patient takes 1 mg of dexamethasone orally at 11 PM and the plasma cortisol level is measured at 8 AM the next morning. An 8-AM plasma corti- sol value of less than 5 µg/dL definitely rules out Cushing’s syndrome. However, the converse is not true. An 8-AM plasma cor- tisol level of more than 5 µg/dL is not diagnostic of Cushing’s syndrome, since several factors continued
History and physical findings suggest Cushing’s syndrome
Conduct overnight dexamethasone (Decadron, Hexadrol) suppression test (1 mg PO at 11 PM)
8-AM plasma cortisol < 5 µg/dL rules out Cushing’s syndrome ’
8-AM plasma cortisol > 5 µg/dL calls for low-dose dexamethasone suppression test (0.5 mg PO q6h for 2 days)
8-AM cortisol < 5 µg/dL rules out Cushing’s syndrome
8-AM cortisol > 5 µg/dL is diagnostic of Cushing’s syndrome; conduct high-dose dexamethasone suppression test (2 mg PO q6h for 2 days)
8-AM cortisol < 5 µg/dL indicates high likelihood of Cushing’s disease (central corticotropin overproduction)
8-AM cortisol > 5 µg/dL indicates a cause other than central corticotropin overproduction
Obtain CT or MRI scan of head
Obtain CT scan of abdomen and adrenal scan; perform adrenal arteriography, adrenal vein catheterization
can lead to false-positive results. For example, the patient may not have taken the dexamethasone dose or may have gotten sick and thrown it up. He or she may be taking a medication, such as phenobarbital or phenytoin (Di- lantin), that stimulates the hepat-
ic cytochrome P-450 system, thus accelerating dexamethasone metabolism. Also, a patient who is under a great deal of psychiat- ric, psychosocial, or physical organic stress may exhibit rela- tive nonsuppressibility of the 8-AM plasma cortisol level. Of
course, patients who truly have Cushing’s syndrome also have an 8-AM plasma cortisol value of more than 5 µg/dL.
If overnight dexamethasone suppression of the plasma corti- sol level fails to occur, it is neces- sary to proceed to the second
Downloaded by [Monash University Library] at 06:28 03 July 2016
The suspicion of corticotropin overproduction by the pituitary should lead to prompt CT or MRI of the head.
stage of screening, ie, the 2-day low-dose suppression test first proposed in 1960 by Dr Grant Liddle at Vanderbilt University, Nashville, Tennessee. The pa- tient takes 0.5 mg of dexametha- sone orally every 6 hours for 2 days, for a total of eight separate doses. The test was previously performed using urinary 17- hydroxycorticosteroid measure- ment as the end point, but it is simpler to obtain a plasma corti- sol value at 8 AM on day 3. Lack of suppression on the low-dose test is definitely diagnostic of Cushing’s syndrome, and then the task is to localize the lesion.
The third stage is the 2-day high-dose dexamethasone sup- pression test. Here, the patient takes a 2-mg (rather than a 0.5- mg) dose of dexamethasone orally every 6 hours for 2 days. Cortisol suppression to less than 5 µg/dL on the high-dose test, but not on the low-dose test, is character- istic of Cushing’s disease. Unfor- tunately, there are some excep- tions: A few patients with Cush- ing’s disease show suppression on the low-dose test, and a few fail to show suppression on the high-dose test. Nonetheless, this sequential test is often a good way to identify patients with Cushing’s disease. In patients with adrenal adenoma, adrenal carcinoma, or ectopic corticotro- pin production, cortisol suppres- sion should not occur even after
the high-dose dexamethasone test, since cortisol secretion is not being driven by corticotropin.
Studies to localize site of lesion
The three stages of dexametha- sone testing theoretically allow the clinician to determine if the pituitary is the primary site of disease. In practice, however, localization of the lesion in Cush- ing’s syndrome can be a very challenging task. Patients in whom pituitary overproduction of corticotropin is suspected should, early on, undergo either CT or MRI of the head. Some- times, however, no apparent structural abnormalities of the pituitary are found on these scans, even in a patient who later proves to have Cushing’s disease.
One test that obviously may help in determining the site of disease is corticotropin radioim- munoassay. Patients with adre- nal adenoma or carcinoma have very low corticotropin levels, since corticotropin secretion is suppressed by excess circulating glucocorticoids. In contrast, pa- tients with Cushing’s disease usually have corticotropin levels that are modestly elevated or at the upper limit of the normal range. Of course, the highest corticotropin levels are found in patients with ectopic corticotro- pin production. Unfortunately, a consistently reliable corticotro-
pin radioimmunoassay is not universally available in the Unit- ed States. Moreover, many prob- lems still prevent the assay from achieving its complete clinical potential. As noted earlier, one complicating factor is the marked diurnal corticotropin variation, 10 with high values during the night and almost undetectable levels by late afternoon. Also, cortico- tropin has a very short half-life in plasma. Because corticotropin is a fragile molecule, blood samples must be drawn into prechilled tubes, kept on ice, and separated immediately when removed from the ice.
CT of the abdomen is the most widely available technique for visualizing the adrenal glands, and it is often quite helpful in identifying adrenal masses. Some investigators have also advocated adrenal scanning using a ra- dioiodinated contrast agent to define the site of disease in Cush- ing’s syndrome. The currently recommended agent is 6-beta- iodomethylnorcholesterol tagged with iodine 131,11 but this is available in only a limited num- ber of medical centers. An adre- nal scan may show enlargement of the involved gland in cases of adrenal adenoma or carcinoma. There should be essentially no uptake by the opposite adrenal because corticotropin is sup- pressed by the hormonal output from the involved gland. When
continued on page 89
Downloaded by [Monash University Library] at 06:28 03 July 2016
Some experts urge that a CT scan of the chest be obtained before pituitary microsurgery so as not to miss malignancies that produce ectopic corticotropin.
the corticotropin level is in- creased, uptake by both glands is usually very prominent.
However, the adrenal scan is often far from definitive, and other localizing procedures may be needed, including adrenal arteriography and adrenal vein catheterization. The latter proce- dure entails obtaining blood samples from each of the two adrenal veins to determine which gland is producing hormone in excess.
One of the newest tests used to detect the specific cause of Cush- ing’s syndrome involves the use of synthetic CRH to elicit an increase in serum cortisol.12 Patients with Cushing’s disease respond to CRH stimulation with an increase in cortisol, whereas patients with other causes of glucocorticoid excess show no such increase. This reflects the pituitary origin of corticotropin excess in Cushing’s disease. The CRH stimulation test may be used more widely once synthetic CRH becomes generally avail- able.
Treatment
Obviously, treatment of Cush- ing’s syndrome depends on the underlying cause. If the disease is the result of increased corticotro- pin secretion by the pituitary, pituitary microsurgery should be the initial therapeutic ap- proach.13.14 Microsurgery is a very
effective means of treating corti- cotropin excess, especially when a microadenoma is demonstrated preoperatively. Of course, the effectiveness of pituitary surgery varies with the skill and experi- ence of the surgeon; major medi- cal centers generally have the largest body of experience. Even in the absence of direct evidence of a microadenoma, a trans- sphenoidal approach is usually justified when laboratory data point toward a central corticotro- pin source. However, according to some authorities, a CT scan of the chest should be obtained before pituitary microsurgery so as not to miss malignancies producing ectopic corticotropin.
When pituitary surgery does not eradicate corticotropin ex- cess, the next approach is usually conventional irradiation of 4,000 to 5,000 rads. The success rate may be disappointingly low, however. 2,15,16 Heavy-particle irradiation, which so far is avail- able in only three cities (San Francisco, Houston, and Bos- ton), 17 yields better results than conventional radiation therapy, but the need for a cyclotron makes its use very impractical for most patients with Cushing’s disease.
Some experts have advocated use of mitotane (Lysodren), also known as o,p’-DDD, to treat intractable Cushing’s disease. 18 This potent metabolic inhibitor
directly blocks the adrenal syn- thesis of steroids. Its side effects include nausea, vomiting, and disordered mentation. Although good results have been achieved in some patients, mitotane thera- py for Cushing’s disease should be used only rarely. Aminoglu- tethimide (Cytadren) and metyra- pone (Metopirone) are equally toxic drugs that have also been used, usually in combination with other agents, to inhibit the adrenal synthesis of steroids.2,19 Bilateral adrenalectomy is yet another approach that is still used in a few centers. Subtotal adrenalectomy places the patient at high risk for postoperative recurrence and thus should not be used.
Some experts have tried to reduce surgical morbidity, which may be increased as a result of poor tissue elasticity and poor wound healing in the presence of excessive glucocorticoids, by administering mitotane preoper- atively. This approach requires that fairly large doses of glucocor- ticoids be given postoperatively to ensure that the patient does not experience an addisonian crisis. Once this is accomplished, rapid tapering must be initiated.
Surgery is clearly the only option for treating adrenal ade- noma and adrenocortical nodular dysplasia. Again, postoperative management must include cover- age with large doses of glucocorti- continued
coids, followed by fairly rapid tapering. A bilateral approach is needed to eradicate nodular dysplasia.
The treatment of adrenal car- cinoma in the absence of de- monstrable metastases is also surgical. Unfortunately, most of these tumors have already me- tastasized by the time they are detected. In this event, therapy is simply palliative. Drugs that inhibit adrenal steroidogenesis (sometimes in fairly large doses) can markedly decrease tumor size in some patients and ame- liorate the features of Cushing’s syndrome in most patients. Un- fortunately, these drugs have not been shown to prolong overall survival.
Similar considerations apply in the patient with an ectopic malig- nant tumor that produces corti- cotropin. If the primary tumor is resectable, surgery is obviously indicated. Postoperative glucocor- ticoid coverage is needed because pituitary secretion of corticotro- pin has been suppressed for some time. Removal of the tumor may lead to a very marked reduc- tion in the adrenal output of glucocorticoids. If the tumor is unresectable, mitotane (or one of the other metabolic inhibitors) can be used. Generally, however, management should focus pri- marily on the primary tumor rather than on Cushing’s syn- drome per se.
Summary
In addition to prolonged glu- cocorticoid therapy (not dis- cussed here), at least five other conditions cause Cushing’s syndrome. They are exces- sive corticotropin secre- tion by the pituitary gland (which results in Cushing’s disease), ectopic production of corticotropin by malignant nonpituitary tumors, benign adrenal adenoma, adrenal car- cinoma, and primary adreno- cortical nodular dysplasia. Each can be distinguished by a specific pathophysiologic pro- cess that triggers the adrenal
glands to overproduce gluco- corticoids.
At present, diagnosis of Cushing’s syndrome or disease relies heavily on the dexa- methasone (Decadron, Hexa- drol) suppression test. After diagnosis, other studies, in- cluding computed tomography, magnetic resonance imaging, and corticotropin radioimmu- noassay, can be used to localize the site of the lesion. Treatment, of course, depends on the un- derlying cause. FCM
Address for correspondence: James V. Felicetta, MD, Veterans Affairs Medical Center, Seventh St and Indian School Rd, Phoenix, AZ 85012.
References
1. Medical Knowledge Self-Assessment Program (MKSAP) VII. Philadelphia: Am College of Physicians, 1988
2. Howlett TA, Rees LH, Besser GM. Cushing’s syndrome. Clin Endocrinol Metab 1985;14(4):911-45
3. Belsky JL, Cuello B, Swanson LW, et al. Cushing’s syndrome due to ectopic production of corticotropin- releasing factor. J Clin Endocrinol Metab 1985;60(3): 496-500
4. Schteingart DE, Lloyd RV, Akil H, et al. Cushing’s syndrome secondary to ectopic corticotropin-releasing hormone-adrenocorticotropin secretion. J Clin Endo- crinol Metab 1986;63(3):770-5
5. Hutter AM Jr, Kayhoe DE. Adrenal cortical carcinoma: clinical features of 138 patients. Am J Med 1966:41(4):572-80
6. Larsen JL, Cathey WJ, Odell WD. Primary adrenocortical nodular dysplasia, a distinct subtype of Cushing’s syndrome: case report and review of the literature. Am J Med 1986;80(5):976-84
7. Jeremy JY, Dandona P. Inhibition by hydrocorti- sone of prostacyclin synthesis by rat aorta and its reversal with RU486. Endocrinology 1986;119(2):661-5 8. Crapo L. Cushing’s syndrome: a review of diagnostic tests. Metabolism 1979:28(9):955-77
9. Aron DC, Tyrrell JB, Fitzgerald PA, et al. Cushing’s syndrome: problems in diagnosis. Medicine (Baltimore) 1981;60(1):25-35
10. Krieger DT. Rhythms in CRF, ACTH, and cortico- steroids. In: Krieger DT, ed. Endocrine rhythms. New York: Raven Press, 1979:123-42
11. Fig LM, Gross MD, Shapiro B, et al. Adrenal
localization in the adrenocorticotropic hormone- independent Cushing syndrome. Ann Intern Med 1988; 109(7):547-53
12. Nieman LK, Chrousos GP, Oldfield EH, et al. The ovine corticotropin-releasing hormone stimulation test and the dexamethasone suppression test in the differential diagnosis of Cushing’s syndrome. Ann Intern Med 1986:105(6):862-7
13. Melby JC. Therapy of Cushing disease: a consen- sus for pituitary microsurgery. Ann Intern Med 1988; 109(6):445-6
14. Tyrrell JB, Brooks RM, Fitzgerald PA, et al. Cushing’s disease: selective trans-sphenoidal resection of pituitary microadenomas. N Engl J Med 1978; 298(14):753-8
15. Orth DN, Liddle GW. Results of treatment in 108 patients with Cushing’s syndrome. N Engl J Med 1971;285(5):243-7
16. Jennings AS, Liddle GW, Orth DN. Results of treating childhood Cushing’s disease with pituitary irradiation. N Engl J Med 1977;297(18):957-62
17. Kjellberg RN, Kliman B. Lifetime effectiveness of a system of therapy for pituitary adenomas, emphasizing Bragg peak proton hypophysectomy. In: Linfoot JA, ed. Recent advances in the diagnosis and treatment of pituitary tumors. New York: Raven Press, 1979:269-78
18. Hutter AM Jr, Kayhoe DE. Adrenal cortical carcinoma: results of treatment with o.p’DDD in 138 patients. Am J Med 1966:41 (4):581-92
19. Orth DN. Metyrapone is useful only as adjunctive therapy in Cushing’s disease. Ann Intern Med 1978; 89(1):128-30