Primary Adrenal Causes of Cushing’s Syndrome
Diagnosis and Surgical Management
ROGER R. PERRY, M.D.,* LYNNETTE K. NIEMAN, M.D.,t GORDON B. CUTLER, JR., M.D.,t GEORGE P. CHROUSOS, M.D.,t D. LYNN LORIAUX, M.D.,t JOHN L. DOPPMAN, M.D.,# WILLIAM D. TRAVIS, M.D.,” and JEFFREY A. NORTON, M.D .*
Cushing’s syndrome is rare with only 20% of patients having a primary adrenal cause of hypercortisolism. We have developed a strategy to evaluate patients with suspected Cushing’s syndrome and to localize the pathologic condition responsible for the hy- percortisolism. This report reviews the last 11 consecutive pa- tients who had a primary adrenal cause of hypercortisolism. Each patient had elevated 24-hour urine free cortisol and 17-hydroxy- corticosteroid excretion consistent with hypercortisolism. All but one patient had undetectable plasma ACTH levels. No patient suppressed urinary steroid levels with high-dose dexamethasone and only one patient increased plasma ACTH or cortisol levels with oCRH, findings that were consistent with a pituitary-in- dependent form of hypercortisolism. No patient had a pituitary tumor detected by computed tomography or magnetic resonance imaging, and eight patients had adrenal tumors accurately im- aged. MRI of the adrenal glands correctly diagnosed adenoma in 5 of 6 patients with adenomas, carcinoma in 1 patient, and ACTH-producing pheochromocytoma in 1 patient. One tumor classified as carcinoma by MRI appeared on pathologic exam- ination to be an adenoma. Three patients underwent petrosal sinus sampling for measurement of ACTH before and after oCRH administration, and each had petrosal sinus ACTH levels equal to peripheral levels, consistent with a primary adrenal cause of hypercortisolism. Two of these patients had typical bilateral pigmented micronodular adrenocortical disease and the third pa- tient had macronodular adrenocortical hyperplasia. Each of the 11 patients was cured of hypercortisolism by unilateral or bi- lateral adrenalectomy and no patient has developed recurrent disease during the 7 to 29 month follow-up period. New modal- ities including the ovine CRH test, MRI, and petrosal sinus sam- pling have improved the evaluation of certain patients with Cushing’s syndrome.
P ATIENTS WITH CUSHING’S SYNDROME have symptoms and signs affecting virtually every organ system. Physician recognition and diagnosis of the syndrome is often delayed because endogenous hyper-
From the Surgical Metabolism Section, Surgery Branch,* and the Laboratory of Pathology,” National Cancer Institute, the Developmental Endocrinology Branch, National Institute of Child Health and Human Development, t and the Department of Radiology of the Clinical Center,# National Institutes of Health, Bethesda, Maryland
cortisolism is rare (10 patients per 1 million),’ the signs of hypercortisolism occur in other disease processes, the changes may be subtle, and no particular symptom or sign occurs in every patient. Without recognition and proper treatment, Cushing’s syndrome results in high rates of morbidity and mortality. Untreated patients have a 5- year survival rate of 50% with most deaths resulting from infection or cerebrovascular disease.2 In the 1950s, the treatment of choice for Cushing’s syndrome was bilateral adrenalectomy, which markedly increased survival com- pared to untreated patients.2 However, it later became clear that most patients with Cushing’s syndrome do not have primary adrenal disease.3 Adrenalectomy is now used principally for patients with primary adrenal disease, oc- cult ectopic ACTH syndrome, or pituitary disease that is refractory to other treatments.
The diagnosis of Cushing’s syndrome requires recog- nition of the clinical syndrome confirmed by appropriate laboratory tests. Both biochemical and radiological studies are necessary to achieve a diagnostic accuracy approaching 100%. Recent advances, including the ovine corticotropin releasing hormone (oCRH) stimulation test,4-7 adrenal magnetic resonance imaging (MRI),8,9 and petrosal sinus sampling,10 have facilitated the preoperative diagnosis of these patients. The strategy for the diagnosis and local- ization of the causes of Cushing’s syndrome that we have employed at the National Institutes of Health (NIH) is outlined in Figure 1. This report will review our last 11 consecutive patients with Cushing’s syndrome due to pri-
Correspondence and reprint requests: Jeffrey A. Norton, M.D., Surgery Branch, National Cancer Institute/NIH, Building 10, Room 2B05, Be- thesda, MD 20892
Establish Hypercortisolism
Overnight Dex. Test (Screening) 24h Urine Free Cortisol 24h Urine 17-OHS/g Creat.
Pituitary- Ectopic- Adrenal
oCAH Test (Includes ACTH)
Low Dose-High Dose Dex. Test
Pituitary
Ectopic
Adrenal
oCRH : ACTH1
Cortisol I
DCAH: ACTH -
Dex .: Low Dose - High Dosel
Cortisol - Low Dose - High Dose —
OCRH: ACTH Undetectable Cortisol -
Dex:
Dex:
Low Dose - High Dose -
Radiologic Localization of Source
CT/MRI Adrenals CT/MRI Sella
Petrosal Sinus Sampling
± Petrosal Sinus Sampling
Pituitary
Ectopic
Adrenal
Petrosal ACTH Gradient Bil. Adrenal Hypertrophy Unil. or Bilateral Macronodules ± Pituitary Adenoma
No Petrosal ACTH Gradient Bil. Adrenal Hypertrophy
ACTH Undetectable Adenoma Carcinoma Micronodular Disease (nt. CT / MRI)
. Pituitary Adenoma vs. Ectopic CRH {rare)
Chest CT/MRI Abdomen CT/MRI Bronchoscopy Needle Aspiration Other Hormones
· Adenoma vs. Carcinoma
Plasma CRH
lodocholesterol Uptake
· Macronodule vs. Autonomous Adrenal Adenoma
( Catecholamines, Calcitonin, Gastrin, VIP, PP, Neurotensin, etc.)
lodocholesterol Uptake
mary adrenal disease, with emphasis on the diagnostic evaluation and the current indications and technical ap- proach to adrenalectomy.
Methods
Biochemical Evaluation
The records of 11 consecutive patients with an adrenal cause of Cushing’s syndrome who were treated between 1985 and 1988 were reviewed. All patients were studied prospectively according to the diagnostic protocol outlined in Figure 1. Hypercortisolism was evaluated by measuring morning and evening plasma cortisol, 24-hour urine free cortisol (UFC), and 24-hour urine 17-hydroxycorticoste- roid (17-OHS) excretion. Pituitary dependence or inde- pendence was assessed with the 6-day low-dose and high- dose dexamethasone suppression test and the oCRH stimulation test.4,5 A positive response of the dexameth- asone test consistent with Cushing’s disease (pituitary etiology) was defined as a decrease of urinary 17-hydroxy-
corticosteroids of ≥ 50% or urinary free cortisol of ≥ 80% of the mean baseline levels on the second day of high- dose dexamethasone administration.
Ovine CRH was given at a dose of 1 mcg/kg iv and a positive response (consistent with Cushing’s disease) was defined as an increase in the mean level of adrenocorti- cotropic hormone (ACTH) or cortisol of more than four times the intra-assay coefficient of variation of the mean baseline concentration. The typical intra-assay coefficient of variation was 6% for baseline ACTH levels between 9 and 30 pg/ml and cortisol levels between 4 and 35 µg/dl. The inter-assay coefficient of variation was 14% to 26% for ACTH and 9% to 12% for cortisol.5 Additional hor- monal tests including 24-hour urine catecholamines, vanillylmandelic acid (VMA), and metanephrines, plasma calcitonin, and other peptide hormone tumor markers were performed, if indicated, to attempt to localize the source of ectopic ACTH secretion (Fig. 1).
Radiological Evaluation
Computed tomography (CT) and/or magnetic reso- nance imaging (MRI) of the adrenal glands and the pi- tuitary gland were performed in all patients. Petrosal sinus sampling was performed in patients in whom initial bio- chemical evaluation indicated a primary adrenal cause of Cushing’s syndrome and in whom CT or MRI either failed to show an adrenal abnormality or indicated bilateral dif- fuse or nodular hyperplasia consistent with Cushing’s dis- ease. Bilateral simultaneous sampling of both petrosal si- nuses and a peripheral vein for plasma ACTH levels was performed before and after oCRH administration.1º Pa- tients with Cushing’s disease generally have a gradient greater than 1.6:1 in ACTH levels on one side compared to the contralateral side and to the periphery. Patients with a primary adrenal etiology have low ACTH levels that are similar in the petrosal sinuses and peripheral veins. In addition, iodocholesterol scans were performed in some patients with an uncertain diagnosis in whom a primary adrenal etiology was suspected but no adrenal abnormality could be imaged by CT or MRI scanning.
As previously reported,8,9 the MRI T2-weighted image intensity of the adrenal tumor was compared to the image intensity of the liver. If the intensity of the adrenal mass was more than three times that of the liver, the tumor was presumptively considered a pheochromocytoma based on MRI criteria. If the adrenal gland image was slightly (approximately 1.5 to 2.5 times) brighter than the liver, the tumor was considered a probable carcinoma, and if the adrenal image was equal to or less bright than the adjacent liver, the tumor was predicted to be a benign adenoma. The prospective diagnoses of adrenal masses based on MRI criteria were subsequently compared to the pathologic diagnoses of the specimens.
Surgery and Pathology
All patients underwent either unilateral or bilateral ad- renalectomy. The posterior approach with resection of the 12th rib was performed for small lesions even if bi- lateral adrenalectomy was necessary. Bilateral adrenal- ectomy was performed through two separate posterior in- cisions. The anterior approach was chosen for all adrenal lesions greater than 6 cm in diameter, or if the MRI in- dicated carcinoma or pheochromocytoma, or if additional abdominal disease requiring surgical treatment was pres- ent. In general, adrenal arteries and veins were controlled by hemoclips and not suture.
Adrenalectomy specimens were examined carefully for cortical and medullary abnormalities. The glands were weighed and examined grossly by a single pathologist for diffuse or localized abnormalities such as nodularity, hy- perplasia or atrophy. These gross abnormalities were cor- related with microscopic and clinical findings resulting in subclassification according to previously described criteria for adrenal adenoma, 11-13 carcinoma,11-13 pheochromo- cytoma,11 diffuse cortical hyperplasia,11 bilateral primary pigmented micronodular adrenocortical disease, 11,14-16 and macronodular hyperplasia.11
Results
The results are grouped according to the final pathologic diagnosis. The four diagnoses were adrenal cortical neo- plasm (adenoma and carcinoma, n = 7), adrenal med- ullary neoplasm (pheochromocytoma, n = 1), bilateral primary pigmented micronodular adrenocortical disease (n = 2), and bilateral adrenocortical macronodular hy- perplasia (n = 1).
Signs and Symptoms
All except one of the 11 patients were female. The mean age of all patients was 39.6 years (range, 14 to 65 years). The two female patients with pigmented micronodular disease were the youngest with a mean age of 17 years (Table 1). Weakness and weight gain (truncal obesity) were the most common presenting symptom and sign, respec- tively, with 8 of 11 patients presenting with each. The remainder of the symptoms and signs are listed in Table 1 in order of decreasing frequency. No single symptom or sign occurred in every patient.
Endocrine Studies
Plasma ACTH levels were low in all patients except one who had an ACTH-producing pheochromocytoma (Table 2). Twenty-four-hour urine free cortisol and 17- hydroxycorticosteroid levels were elevated in all 11 pa- tients. The oCHR test did not increase plasma ACTH or cortisol levels significantly in any patient except one,
| Pathologic Group | n | Mean Age (range) | Female | Weakness | Obesity | Hirsuitism | Hypertension | Amenorrhea | Mental Changes | Edema | Glucose Intolerance |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Adrenal Cortical Neoplasm | 7 | 43 (26-65) | 7 (100%)* | 5 (71%) | 5 (71%) | 3 (43%) | 4 (57%) | 3 (43%) | 4 (57%) | 3 (43%) | 2 (29%) |
| Adrenal Medullary Neoplasm | 1 | 38 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 |
| Adrenal Cortical Micronodular Disease | 2 | 17.5 (14-21) | 2 (100%) | 2 (100%) | 2 (100%) | 2 (100%) | 0 (0%) | 1 (50%) | 0 (0%) | 0 (0%) | 0 (0%) |
| Adrenal Cortical Macronodular Hyperplasia | 1 | 61 | 0 | 0 | 1 | NA | 1 | NA | 0 | 0 | 0 |
| Total | 11 | 39.6 (14-65) | 10 (91%) | 8 (73%) | 8 (73%) | 6 (60%) | 6 (55%) | 5 (50%) | 5 (46%) | 4 (36%) | 3 (27%) |
* Number in parentheses indicates percentage of patients who were female or the percentage of patients who presented with a given sign or symptom.
NA, not applicable.
| Pathologic Group | n | Plasma ACTH pg/ml | Per cent Delta ACTH with oCRH | Morning Plasma Cortisol mcg/dl | Per cent Delta Cortisol with oCRH | Urine-free cortisol mcg/24 h | Per cent Delta UFC with High- Dose Dexamethasone | Urine 17-OHS mg/gr creatine/ 24 h | Per cent Delta 17-OHS with High-Dose Dexamethasone |
|---|---|---|---|---|---|---|---|---|---|
| Reference range | <4-25* | > +25%+ | 4-20* | > +25%+ | 20-95* | -80 to -100%} | 2-6* | -50 to -100%} | |
| Adrenal cortical neoplasm | 7 | <4 (2.6, 5.1) | +7.5% (0, +197) | 22.4 (10.4. 24.0) | -2% (-16, +87) | 314 (94, 873) | +29% (-74, +78) | 16.8 (9.2, 20.7) | -2% (-46, +36) |
| Adrenal medullary neoplasm | 1 | 139 | -15% | 106 | +19 | 13,700 | -34% | 150 | -35% |
| Adrenal cortical micronodular disease | 2 | <4 | 0% | 18.5 (18.4, 18.6) | +1% (-2, +4) | 272 (195, 349) | +73% (+46, +100) | 11.5 (9.5, 13.5) | +42% (+33, +51) |
| Adrenal cortical macronodular hyperplasia | 1 | <4 | 0% | 20.1 | -16% | 202 | Not done | 23.6 | +5% |
All numbers for groups with n > 1 are listed as medians with the range given in parentheses.
* Normal range.
Response seen in patients with Cushing’s disease. Pituitary-inde- pendent forms of Cushing’s Syndrome should have less stimulation.
which is consistent with the diagnosis of a primary adrenal cause of Cushing’s syndrome in 9 of 10 patients and ec- topic ACTH syndrome in the final patient. High-dose dexamethasone did not suppress urine free cortisol and 17-hydroxycorticosteroid levels in any patient, which is consistent with a nonpituitary cause of Cushing’s syn- drome in all 11 patients (Table 2). The patient with an ACTH-producing pheochromocytoma had elevated 24- hour urine VMA, metanephrines, and catecholamines.
Radiologic Studies
Preoperative CT of the sella turcica was normal in all patients (Table 3). CT of the adrenal glands was abnormal in nine patients and normal in the two patients with mi- cronodular hyperplasia. The seven patients with adrenal cortical neoplasms each had a unilateral adrenal mass with a normal or atrophic contralateral adrenal gland. The one patient with an ACTH-producing pheochromocytoma had bilateral adrenal hyperplasia visible on CT (Fig. 2) and T2-weighted MRI demonstrated a central area of in- creased signal in the right gland consistent with a pheo- chromocytoma. This lesion was not suspected clinically or on CT exam. The patient with macronodular adrenal hyperplasia had bilateral massive nodular enlargement of both adrenal glands on CT. CT documented the size and location of the adrenal pathologic condition in 82% of the patients with no false-positive results.
Magnetic resonance imaging T1-weighted images gave results similar to CT. It identified the size and location of the adrenal pathologic condition in 82% of patients
# Range for patients with Cushing’s Disease. Pituitary-independent forms of Cushing’s Syndrome should have less suppression. Normal sub- jects have essentially complete suppression (to the assay detection limit) with high-dose dexamethasone.
with no false-positive results. In addition, the T2-weighted image accurately predicted the final pathologic diagnosis in 7 of 8 (88%) tumors, including a pheochromocytoma (Fig. 2), a carcinoma (Fig. 3), and 5 of 6 adenomas. How- ever, T2-weighted MR image predicted a carcinoma in one patient who appears to have an adrenal adenoma (Table 3).
The patients with primary bilateral adrenocortical dis- ease (2 micronodular and 1 macronodular) each under- went petrosal sinus sampling for ACTH levels with and without oCRH to document that no pituitary source of increased ACTH was present. In each patient petrosal sinus ACTH levels were bilaterally less than the detection limit of 4 to 6 pg/ml and no change in levels was elicited following oCRH. In addition, iodocholesterol scans were
performed in these patients with scanning attempted 48 hours after administration of the iodocholesterol. Two patients (one with pigmented micronodular disease and one with macronodular hyperplasia) had uptake in both adrenal glands, but the other patient with pigmented mi- cronodular disease had no accumulation of iodocholes- terol in either adrenal gland.
Surgery and Pathologic Condition
A unilateral adrenalectomy was performed in eight pa- tients and bilateral adrenalectomy in three patients. All patients were cured of endogenous hypercortisolism. Pa- tients undergoing unilateral adrenalectomy were main- tained after operation on maintenance doses of hydro- cortisone (12 to 15 mg/m2/day) until the cortisol response
| Pathologic Group | n | Number with Abnormal Adrenal CT (%) | Number with Abnormal Pituitary CT (%) | Number with Abnormal Adrenal MR (%) | Number in whom MR T2 Ratio Predicted Pathologic Diagnosis (%) |
|---|---|---|---|---|---|
| Adrenal cortical neoplasm | 7 | 7 (100) | 0 (0) | 7 (100) | 6 (86) |
| Adrenal medullary neoplasm | 1 | 1 | 0 | 1 | 1 |
| Adrenal cortical micronodular disease | 2 | 0 (0)* | 0 (0) | 0 (0)* | 0 (0) |
| Adrenal cortical macronodular hyperplasia | 1 | 1 | 0 | 1 | 1 |
* CT and MRI were normal in these 2 patients but iodocholesterol scan indicated hyperfunction of both adrenals in one patient.
to ACTH was normal (6 to 12 months). Patients under- going bilateral adrenalectomy were maintained after op- eration on maintenance doses of hydrocortisone and mineralocorticoid (fluorinef R 50 to 100 mcg/day). All patients had immediate postoperative hydrocortisone doses rapidly tapered so that maintenance doses were reached by postoperative days 5 to 7. The posterior ap- proach was chosen in five patients for either small adrenal cortical adenomas or micronodular adrenal disease. The mean weight of the adrenal adenomas resected through the back was 12 grams and the four resected micronodular disease glands had a mean weight of 5 g (Table 4). The anterior or flank approach was chosen in six patients and each patient had either large hyperplastic adrenal glands or large adrenal neoplasms. Glands removed anteriorly included a carcinoma and a pheochromocytoma, which were larger than glands removed via the posterior ap- proach (Table 4). There were no complications except for one patient who suffered a temporary axillary nerve injury with dominant extremity paresis.
The final pathologic diagnosis was adrenal cortical ad- enoma in 6 patients, pheochromocytoma with associated diffuse adrenocortical hyperplasia in 1 patient, adrenal cortical carcinoma in 1 patient, bilateral primary pig- mented micronodular adrenocortical disease in 2 patients, and bilateral adrenocortical macronodular hyperplasia in 1 patient (Table 4). The three patients with bilateral nod- ular adrenal cortical abnormalities deserve special men- tion. The patient with macronodular hyperplasia had markedly thickened adrenal cortices with multiple nodules ranging in size from several millimeters to 4 cm. The total weight of both glands combined was 95 g. Histologically the cortical macronodules were composed of large cells with clear to granular cytoplasm alternating with compact eosinophilic cells. Both patients with pigmented micro- nodular disease had small to normal-sized adrenal glands with a combined weight of 9.5 g in one and 9.4 g in the other. The cross-sectioned surface of each gland revealed multiple brown to gray 1 mm to 2 mm pigmented nodules. Histologically these nodules were circumscribed but not encapsulated and consisted of large cells with darkly eo- sinophilic cytoplasm containing varying amounts of
brown cytoplasmic pigment. Occasional foci of lipoma- tous change were seen in the nodules. The intervening adrenal cortex among these nodules was atrophic.
Discussion
The diagnostic evaluation of patients with Cushing’s syndrome has advanced during the past 5 years. New tests such as the ovine CRH test,4-7 petrosal sinus sampling for ACTH levels,10 and magnetic resonance imaging of the adrenal glands8,9 have contributed to the ability to achieve an accurate diagnosis. These tests have been incorporated into our diagnostic protocol for patients with Cushing’s syndrome (Fig. 1, Table 5) and are currently undergoing prospective evaluation in a large series of patients.
Most patients (75%) who present with Cushing’s syn- drome have Cushing’s disease due to an ACTH-producing pituitary adenoma.1 Often these pituitary adenomas are microadenomas and are not seen on high resolution CT.17 Bilateral petrosal sinus sampling for ACTH in the patient with Cushing’s disease will usually demonstrate a gradient and thereby localize the tumor to one half of the pituitary gland, enabling the neurosurgeon to successfully resect it.10 In addition, petrosal sinus sampling can provide di- agnostic information in the patient with ectopic ACTH syndrome or primary nodular adrenocortical disease whose condition is confusing, because patients with Cushing’s disease have a clear ACTH gradient between the petrosal sinus and the peripheral veins, whereas pa- tients with ectopic ACTH syndrome or primary nodular adrenocortical disease do not.18
The present study excluded patients with Cushing’s disease (pituitary adenoma) and patients with ectopic ACTH syndrome from nonadrenal sources and focused only on patients with a primary adrenal cause of Cushing’s syndrome. Each patient in this study had clear biochem- ical evidence for hypercortisolism (elevated 24-hour urine free cortisol and 17-hydroxycorticosteroid excretion; Ta- ble 2). In addition, no patient in this study had normal suppression of urinary steroid levels following the oral administration of low-dose dexamethasone, again consis- tent with Cushing’s syndrome. High-dose dexamethasone
2
A
B
failed to suppress urinary free cortisol levels or urinary 17-OHCS levels by ≥ 80% or 50%, respectively, in all 11 patients, and oCRH administration did not increase plasma ACTH or cortisol levels by 25% in 10 of the 11 patients (Table 2). Thus, these indirect tests were useful in excluding pituitary-dependent Cushing’s disease, but the oCRH test lacked 100% diagnostic accuracy for this purpose. ACTH levels were low in all but one patient and thus were also consistent with a primary adrenal cause of
hypercortisolism in 91% of the patients. Finally, pituitary imaging studies failed to demonstrate an abnormality in any patient (Table 3). However, this information is of limited diagnostic usefulness because approximately 60% of patients with Cushing’s disease will have microadeno- mas that are not seen with CT.17
Patients with endocrinologic evidence for a primary adrenal source of Cushing’s syndrome (hypercortisolism, low ACTH levels, failure of suppression with high-dose
C
dexamethasone, and failure of stimulation with oCRH) have, in most cases, a solitary adrenal mass (as in 7 of the 11 patients in this study). With the above endocrinological evidence and a clear unilateral abnormality on CT or MRI, the evaluation is complete. The only remaining question for the surgeon is what type of incision to make. The MRI scan can help answer that question. It can pro- vide evidence for a carcinoma (Fig. 3) that requires the anterior or thoracoabdominal approach to allow complete abdominal exploration and en-bloc tumor resection. The posterior approach is preferred for small adrenal cortical masses (< 6 cm) or micronodular adrenals (Table 4), es- pecially because patients with Cushing’s syndrome usually have truncal obesity, which can hinder the abdominal approach. The adrenal glands are more readily accessible in these patients via the posterior approach and there are less incidences of perioperative morbidity. However, gen- eral access is limited with the posterior approach and with larger tumors (≥ 6 cm) we would ordinarily select the anterior or flank approach.
One patient had ectopic ACTH syndrome due to an ACTH-secreting adrenal pheochromocytoma.19,20 We recommend measurement of urinary catecholamines in patients with suspected ectopic ACTH syndrome (Fig. 1). Patients with ectopic ACTH syndrome often have the most severe degree of Cushing’s syndrome. This 38-year- old woman had plasma and urinary cortisol levels much greater than the other patients (Table 2) and such severe weakness that she was almost bedridden. CT and MRI documented bilateral adrenal enlargement with an adrenal mass on one side and contralateral diffuse enlargement (Figs. 2A and B). The MRI T2-weighted image demon- strated marked brightness of the adrenal mass consistent
with a pheochromocytoma (Fig. 2C). After preoperative preparation with phenoxybenzamine and propranolol, the pheochromocytoma was resected through an abdominal approach.
The final three patients with bilateral nodular adre- nocortical disease were the most challenging diagnosti- cally. As stated before, endocrinological evaluation of these patients was consistent with a primary adrenal cause of hypercortisolism (Table 2). However, CT and MRI ex- aminations depicted normal adrenal glands in two patients and diffuse macronodular enlargement of both glands in the final patient. Petrosal sinus sampling was performed in each patient and levels of ACTH were low, consistent with a pituitary-independent cause of hypercortisolism. In young female patients who may have bilateral pig- mented micronodular adrenal disease, we recommend iodocholesterol scans to attempt to demonstrate that the cortisol excess is coming from the adrenal glands as op- posed to ectopic adrenal rest tissue or a factitious source (Fig. 1). Iodocholesterol scans were performed in both micronodular patients, but one patient failed to have in- creased uptake in the adrenal glands. Nevertheless, adrenal glands resected from both patients had bilateral pigmented micronodular adrenal disease (Table 4) identical to the previous descriptions of that pathologic entity as a primary adrenal cause of hypercortisolism.11,14-16 Recent obser- vations support the hypothesis that hypercortisolism in pigmented micronodular disease may be due to circulating immunoglobulins that stimulate adrenal steroidogenesis.21
The patient with macronodular adrenocortical hyper- plasia is the most problematic from an etiologic stand- point. This adrenal abnormality is usually associated with a long-standing ACTH-secreting pituitary microade-
A
B
noma.11,22 It may be that this patient had a transition from Cushing’s disease to autonomous cortisol-secreting adrenal macronodules that suppressed peripheral ACTH levels to low or undetectable levels. This explanation has been proposed recently for a similar patient.23 Our patient
is now being closely followed for evidence of an emerging pituitary adenoma:
In summary, a diagnostic strategy for the evaluation of patients with Cushing’s syndrome has been presented (Fig. 1) with the focus on patients with primary adrenal causes
C
of endogenous hypercortisolism. The experience includes patients with adrenal cortical tumors (both adenoma and carcinoma), adrenal medullary tumors producing ACTH, and bilateral nodular adrenocortical disease. Each patient was cured of cortisol excess by unilateral or bilateral ad- renalectomy and no patient has developed recurrent hy- percortisolism, although follow-up is short (mean, 14.6 months; range 7 to 29 months). The results of this study demonstrate that significant changes have occurred in the
diagnostic work-up of patients with Cushing’s syndrome. New modalities including the ovine CRH test, petrosal sinus sampling, and magnetic resonance imaging each may add more certainty to the evaluation of these difficult patients. Because these studies do increase cost, all studies may not be indicated in all patients. Patients with primary adrenal tumors need only the establishment of hypercor- tisolism and an imaging study such as CT or MRI. How- ever, patients with primary bilateral adrenocortical disease
| Pathologic Group | n | Number with Posterior Approach (%) | Pathology of Specimens, Posterior Approach | Mean Weight of Gland, Posterior Approach (g) | Pathology of Specimens, Anterior Approach | Mean Weight of Gland, Anterior Approach (g) |
|---|---|---|---|---|---|---|
| Adrenal cortical | 7 | 3 (43) | Adenoma | 12 | 1 cancer | 167 |
| neoplasm | 3 adenomas | 33 | ||||
| Adrenal medullary neoplasm | 1 | 0 | None | Pheochromocytoma with diffuse hyperplasia of adjacent cortex | 30 | |
| Adrenal cortical micronodular disease | 2 | 2 (100) | Bilateral pigmented 1-2 mm nodules | 5 | None | |
| Adrenal cortical macronodular hyperplasia | 1 | 0 | None | Bilateral macronodular cortical hyperplasia (3-4 cm diameter nodules) | 47 |
| Serum ACTH | oCRH Test | High-Dose Dexamethasone Test | CT/MRI Sella | CT/MRI Adrenals | Petrosal Sinus Sampling | Iodocholesterol Scan | |
|---|---|---|---|---|---|---|---|
| Pituitary Cushing's | Normal to mildly elevated | + ☒ | >50% suppression | +/- | Mild bilateral enlargement or normal | + ☒ | + ☒ |
| Ectopic ACTH | Normal to markedly elevated | - | No suppression | - | Enlarged bilaterally or normal | - ☐ | + ☒ |
| Adrenal neoplasms | Undetectable | - | No suppression | - | Unilateral adrenal mass | NA | Adenomas + Carcinomas - |
| Micronodular disease | Undetectable | - | No suppression | - | Minimal diffuse enlargement ("knobby") or normal | - | +/ -* |
* At 48 hours. May become positive if studied for longer intervals.
+, positive test; - , negative test; +/-, test may be positive or negative. NA, not applicable.
can benefit from the oCRH test and petrosal sinus sam- pling to verify the diagnosis and allow appropriate surgical therapy.
References
1. Loriaux DL, Cutler GB. Diseases of the adrenal glands. In Kohler PO, ed. Clinical Endocrinology. New York: John Wiley and Sons, 1986:167-238.
2. Plotz CM, Knowlton AI, Ragan C. The natural history of Cushing’s syndrome. Amer J Med 1952;13:597-614.
3. Orth DN, Liddle GW. Results of treatment in 108 patients with Cushing’s syndrome. N Engl J Med 1971;285:243-247.
4. Chrousos GP, Schulte HM, Oldfield EH, et al. The corticotropin- releasing factor stimulation test. N Engl J Med 1984;310:622- 626.
5. Nieman LK, Chrousos GP, Oldfield EH, et al. The ovine cortico- tropin-releasing hormone stimulation test and the dexamethasone suppression test in the differential diagnosis of Cushing’s syn- drome. Ann Intern Med 1986;105:862-867.
6. Chrousos GP, Schuermeyer TH, Doppman J, et al. Clinical appli- cations of corticotropin-releasing factor. Ann Intern Med 1985;102:344-358.
7. Gold PW, Loriaux DL, Roy A, et al. Responses to corticotropin- releasing hormone in the hypercortisolism of depression and Cushing’s disease. Pathophysiologie and diagnostic implications. N Engl J Med 1986;314:1329-1335.
8. Reinig JW, Doppman JL, Dwyer AJH, et al. Adrenal masses dif- ferentiated by MR. Radiology 1986;158:81-84.
9. Doppman JL, Reinig JW, Dwyer AJ, et al. Differentiation of adrenal masses by magnetic resonance imaging. Surgery 1987;102:1018- 1026.
10. Oldfield EH, Chrousos GP, Schulte HM, et al. Pre-operative later- alization of ACTH-secreting pituitary microadenomas by bilateral and simultaneous inferior petrosal venous sinus sampling. N Engl J Med 1985;312:100-103.
11. Page DL, DeLellis RA, Hough AJ. Tumors of the Adrenal, Atlas of
Tumor Pathology, Second Series, Fascicle 23, Armed Forces In- stitute of Pathology, 1986.
12. Hough AJ, Hollifield JW, Page DL, Hartmann WH. Prognostic fac- tors in adrenal cortical tumors. A mathematical analysis of clinical and morphologie data. Am J Clin Pathol 72:390-399, 1979.
13. Weiss LM. Comparative histologic study of 43 metastasizing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 8: 163-169, 1984.
14. Shenoy BV, Carpenter PC, Carney JA. Bilateral primary pigmented nodular adrenalcortical disease. Rare cause of the Cushing syn- drome. Am J Surg Pathol 8:335-344, 1984.
15. McArthur RB, Bahn RC, Hayles AB. Primary adrenocortical nodular dysplasia as a cause of Cushing’s syndrome in infants and children. Mayo Clin Proc 1982;57: 58-63.
16. Larsen JL, Cathey WJ, Odell WD. Primary adrenocortical nodular dysplasia, a distinct sybtype of Cushing’s syndrome. Case report and review of the literature. Am J Med 1986;80:976-984.
17. Saris SC, Patronas NJ, Doppman JL, et al. Cushing syndrome: pi- tuitary CT scanning. Radiology 1987;162:775-777.
18. Zovickian J, Oldfield EH, Doppman JL, et al. Usefulness of inferior petrosal sinus venous endocrine markers in Cushing’s disease. J Neurosurg 1988;68:205-210.
19. Spark RF, Connolly PB, Gluckin DS, et al. ACTH secretion from a functioning pheochromocytoma. N Engl J Med 1979;301:416- 418.
20. Jessop DS, Cunnah D, Millar JG, et al. A pheochromocytoma pre- senting with Cushing’s syndrome associated with increased con- centrations of circulating corticotrophin-releasing factor. J En- docrinol 1987;113:133-138.
21. Wulffratt MM, Drexhage HA, Wiersinga WM, et al. Immunoglob- ulins of patients with Cushing’s syndrome due to pigmented ad- renocortical micronodular dysplasia stimulate in vitro steroido- genesis. J Clin Endocrinol Metab 1988;66:301-7.
22. Doppman JL, Miller DL, Dwyer AJ, et al. Macronodular adrenal hyperplasia in Cushing disease. Radiology 1988;166:347-352.
23. Hermus AR, Pieters GF, Smals AG, et al. Transition from pituitary- dependent to adrenal-dependent Cushing’s syndrome. N Engl J Med 1988;318:966-970.