CLEVELAND CLINIC EXPERIENCE WITH ADRENAL CUSHING’S SYNDROME
JAMES A. DAITCH, DAVID A. GOLDFARB AND ANDREW C. NOVICK From the Department of Urology, Cleveland Clinic Foundation, Cleveland, Ohio
ABSTRACT
Purpose: Cushing’s syndrome due to adrenal adenoma or adrenocortical carcinoma is rare. To understand better the clinical and biochemical presentation of this disorder, as well as therapy efficacy and patient survival, we conducted a retrospective review.
Materials and Methods: Between August 1971 and April 1994, 40 patients presented to our institution with adrenal Cushing’s syndrome (27 adenomas and 13 carcinomas). These groups were analyzed with respect to clinical signs and symptoms preoperatively and postoperatively, biochemical analysis, length of postoperative steroid replacement therapy, disease recurrence and patient survival. Followup was obtained by chart review and telephone interviews and averaged 59.6 ± 66.4 and 47.6 ± 56.2 months for adenoma and carcinoma patients, respectively.
Results: Women predominated in both groups (26 of 27 adenomas, 11 of 13 carcinomas), and tumors affected the left adrenal gland more frequently (19 of 27 adenomas, 9 of 13 carcinomas). Adenoma patients were younger than carcinoma patients (39.6 + 14.4 versus 51.5 ± 16.6 years, p = 0.026) and presented with smaller tumors (3.3 ± 1.0 versus 8.6 ± 4.5 cm., p = 0.001). There was a trend toward increased incidence of glucose intolerance among carcinoma patients but no significant differences in clinical signs or symptoms between adenoma and carcinoma patients could be made. Similarly, while there was no significant difference in biochemical evaluation of adenoma versus carcinoma patients, 24-hour urinary free cortisol and serum lactate dehydroge- nase levels tended to be higher among carcinoma patients. In addition 17-ketosteroid and dehydroepiandrosterone sulfate levels were more elevated in carcinoma than in adenoma pa- tients, and several adenoma patients actually had subnormal levels. Among adenoma patients mean length of steroid replacement therapy was 16.8 ± 9.1 months. However, 7 adenoma patients (25.9%) required greater than 24 months of exogenous steroids, and only 1 of these patients was subsequently weaned off steroid replacement. There were no recurrences among adenoma patients, although there was 1 perioperative death due to hypoglycemia. Ten (76.9%) carcinoma patients had recurrences at a mean followup of 33 months. The 3 and 5-year survival rates were 41.5 and 31.2%, respectively.
Conclusions: While presenting signs and symptoms and hormonal analysis may suggest benign or malignant disease, only tumor size and patient age are reliable preoperative indicators of adrenal adenoma versus adrenocortical carcinoma among patients with adrenal Cushing’s syn- drome. Surgery is curative for adenoma patients, but lifelong steroid replacement may be required. Survival remains poor among carcinoma patients.
KEY WORDS: Cushing’s syndrome; carcinoma, adrenal cortical; steroids
Cushing’s syndrome is characterized by excessive serum glucocorticoids that have escaped regulatory control. Only in a minority of patients is Cushing’s syndrome due to adrenal neoplasms, affecting fewer than 1 per million population per year.1-5 Due to their rarity there is a paucity of data address- ing the natural history of these functional tumors. We were interested in establishing criteria to delineate adrenal ade- nomas from carcinomas preoperatively, as well as defining long-term postoperative outcome among these patients.
MATERIALS AND METHODS
Between August 1971 and April 1994, 40 patients pre- sented to our institution with signs and symptoms of Cushing’s syndrome due to functional adrenal adenoma or adrenocortical carcinoma. Cases were identified by searching
Accepted for publication June 6, 1997.
Editor’s Note: This article is the second of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the ques- tions on pages 2274 and 2275.
the records of patients with Cushing’s syndrome who under- went adrenalectomy. Patients with suspected Cushing’s syn- drome were evaluated with a series of biochemical tests (see figure). Our diagnostic approach has evolved over the last 25 years. We obtain a 24-hour urinary free cortisol. In equivocal cases this is repeated, or a low dose dexamethasone suppres- sion test is performed (1 mg. orally at 11 p.m. or 0.5 mg. every 6 hours for 2 days). Patients with Cushing’s syndrome will not suppress cortisol levels after low dose dexamethasone administration (positive low dose dexamethasone suppres- sion test). Next, several serum adrenocorticotropic hormone (ACTH) levels are obtained. Low or undetectable values are invariably associated with primary adrenocortical disease, and computerized tomography (CT) or magnetic resonance imaging (MRI) of the abdomen is performed. Normal or mod- erately elevated ACTH levels usually indicate pituitary ade- noma, while significantly elevated ACTH levels (greater than twice normal) typically result from ectopic ACTH production. However, to differentiate these 2 entities further, patients with ACTH dependent Cushing’s syndrome should undergo
24-Hour Urinary Free Cortisol / Low Dose DST Plasma ACTH X 2
Low
Normal - Elevated
≥ 2X Normal
CT Adrenal MRI Adrenal
MRI Pituitary High Dose DST Petrosal Sinus Sampling
Adrenal Adenoma Adrenal Carcinoma
Pituitary Adenoma Pituitary Hyperplasia
Ectopic ACTH producing tumor
Algorithm for evaluating patients with suspected Cushing’s syndrome. DST, dexamethasone suppression test. ACTH, adrenocorticotropic hormone.
MRI of the head. Those patients with equivocal MRI then undergo the high dose dexamethasone suppression test (8 mg. orally once or 2 mg. orally every 6 hours for 2 days). Patients with pituitary lesions will suppress cortisol after high dose dexamethasone administration (positive high dose dexamethasone suppression test), whereas those with ectopic ACTH production or adrenal Cushing’s syndrome will fail to suppress cortisol with the high dose dexamethasone suppres- sion test (negative high dose dexamethasome suppression test). If the diagnosis remains in doubt, bilateral petrosal venous sinus sampling before and after stimulation with corticotropin releasing hormone is performed to differentiate pituitary Cushing’s syndrome from other causes.6 In our study all patients in whom data were available had elevated 24-hour urinary free cortisol levels, low serum ACTH levels, and negative high dose dexamethasone suppression test.
Tumors were classified by location, size at excision, and as adrenal adenomas or adrenocortical carcinomas. Pathologi- cal classification was based on light microscopy. Carcinomas were staged according to the system described by Nader et al.7 Localized disease (stage 1) refers to tumor confined to the adrenal gland with no evidence of local or distant spread. Regional disease (stage 2) involves direct extension of tumor into adjacent tissues or organs, or spread to regional lymph nodes. Distant metastasis (stage 3) involves tumor spread to distant organs, the central nervous system, or bones. All patients diagnosed with carcinoma received postoperative 1,1-dichlorodiphenyldicloroethane as tolerated and were con- tinued on exogenous steroid replacement therapy. Adenoma patients were monitored for recovery of the hypothalamic- pituitary-adrenal axis every 3 to 6 months using the me- tyrapone test or ACTH stimulation test, and accordingly weaned from exogenous steroid dependency.
Patients were classified by the signs and symptoms of Cushing’s syndrome at presentation, age, gender and hor- monal evaluation. Recorded physical examination findings included buffalo hump, muscle wasting, acne, peripheral edema, striae, hirsutism, increased weight, easy bruisability, oligomenorrhea, hypertension (systolic blood pressure greater than 140 mm. Hg, diastolic blood pressure greater than 90 mm. Hg, or new requirement for antihypertensive medication within 6 months), and glucose intolerance (de- fined as abnormal glucose tolerance test, necessity for dietary glucose control, or requirement for hypoglycemic agents). Substances assayed included plasma cortisol, 24-hour uri- nary free cortisol, ACTH, 17-hydroxycorticosteroids (17-OH), 17-ketosteroids (17-KS), dehydroepiandrosterone sulfate (DHEA-S), testosterone, aldosterone, and lactate dehydroge- nase. Patient weight, blood pressure and glucose intolerance were compared preoperatively and postoperatively. Length of postoperative steroid replacement therapy, tumor recur-
rence, and patient survival data were obtained by medical record review and by contacting surviving patients. Mean posttreatment followup was 59.6 ± 66.4 months for adenoma patients, and 47.6 ± 56.2 months for carcinoma patients.
Comparisons between adenoma and carcinoma groups were made using the chi-square test or Fisher’s exact test for categorical variables, and applying Student’s t test for con- tinuous variables. Actuarial analysis of tumor recurrence and patient survival rates was determined by the Kaplan- Meier method.8 For all statistical analyses p <0.05 was con- sidered significant.
RESULTS
Age, gender, tumor type, location, size and stage. Of the patients 27 (67.5%) were diagnosed with adenomas and 13 (32.5%) with carcinomas. Adenoma patients presented at a younger age and with smaller tumors. Mean age at diagnosis was 39.6 ± 14.4 years for adenoma patients, compared to 51.5 + 16.6 years for carcinoma patients (p = 0.026). Adeno- mas ranged from 2.0 to 5.0 cm. in greatest diameter (mean 3.3 ± 1.0 cm.), while carcinomas were 3.0 to 15.0 cm. in size (mean 8.6 ± 4.5 cm., p = 0.001). Women predominated in both groups, constituting 26 of 27 adenoma patients (96.3%), and 11 of 13 carcinoma patients (84.6%). The left adrenal gland harbored tumor in the majority of patients, accounting for 19 adenoma cases (70.4%) and 9 carcinoma cases (69.2%). No patient suffered bilateral functional tumors. Among the 13 carcinoma tumors 2 (15.4%) were stage 1, 9 (69.2%) stage 2, and 2 (15.4%) stage 3 at presentation.
Physical findings. All 40 patients manifested clinical find- ings of excess glucocorticoid production. Incidence of various signs and symptoms of Cushing’s syndrome at presentation is summarized in table 1. Although there was no significant
| No. Sign Present/No. Sign Mentioned (%) | |||
|---|---|---|---|
| Adenoma | Ca | Adenoma vs. Ca. (p value) | |
| Buffalo hump | 8/9 (89) | 5/5 (100) | 0.99 |
| Muscle wasting | 11/12 (92) | 11/11 (100) | 0.99 |
| Acne | 7/8 (88) | 5/5 (100) | 0.99 |
| Edema | 9/13 (69) | 5/7 (71) | 0.99 |
| Striae | 13/17 (77) | 4/6 (67) | 0.63 |
| Hirsutism | 15/16 (94) | 8/8 (100) | 0.99 |
| Weight gain | 16/19 (84) | 9/10 (90) | 0.99 |
| Easy bruising | 14/16 (88) | 6/8 (75) | 0.58 |
| Oligomenorrhea | 11/12 (92) | 4/4 (100) | 0.99 |
| Hypertension | 24/27 (89) | 10/11 (91) | 0.99 |
| Glucose intolerance | 9/24 (38) | 7/10 (70) | 0.13 |
difference in any category between adenoma and carcinoma patients, nearly twice as many carcinoma patients suffered glucose intolerance at presentation when compared to ade- noma patients. In addition buffalo hump, muscle wasting, acne, hirsutism and oligomenorrhea were present in all car- cinoma patients for whom these signs were recorded.
Biochemical evaluation. All patients in whom data were available presented with elevated 24-hour urinary free cor- tisol and 17-OH levels, and low/normal ACTH levels. Despite the wide range of hormones assayed, there was no significant difference in any hormonal abnormalities exhibited by ade- noma and carcinoma patients. However, 24-hour urinary free cortisol, 17-KS, DHEA-S and lactate dehydrogenase levels tended toward higher elevation among carcinoma patients than adenoma patients (tables 2 and 3). In addition, among 19 adenoma patients for whom data were available, 13 (68.4%) demonstrated pure biochemical syndrome (glucocor- ticoid excess without elevation of either 17-KS, DHEA-S, testosterone or aldosterone levels). Only 1 of 13 carcinoma patients (7.7%) demonstrated pure biochemical syndrome.
Radiographic evaluation. Results of CT, MRI and angio- gram were available for several patients. Among 20 abdom- inal CTs recorded for adenoma patients, none showed central calcification of the adrenal mass, whereas 4 CTs of 8 carci- noma patients (50%) demonstrated central calcification of the adrenal masses. MRI evaluation revealed increased T2 signal intensity for 2 of 4 adenoma patients (50%) and 3 of 4 carcinoma patients (75%). Finally, 7 of 10 adenoma tumors (70%) displayed increased vascularity on angiogram, whereas 4 of 8 (50%) carcinoma tumors demonstrated hyper- vascularity.
Postoperative changes. After tumor excision, hypertension improved in 100% of adenoma patients and in 77.8% of car- cinoma patients (p = 0.068) as measured by decrease in number of antihypertensive medications necessary to main- tain normotensive systolic and diastolic arterial pressures on greater than 2 postoperative examinations. Average weight loss after surgery was 22.0 ± 16.5 pounds for adenoma pa- tients, and 25.5 + 21.9 pounds for carcinoma patients (p = 0.58). Glucose intolerance as measured by need for diet con- trol, oral hypoglycemics or insulin resolved in 100% of ade- noma patients and in 80% of carcinoma patients (p = 0.12).
Length of steroid dependence. All carcinoma patients re- mained on exogenous steroids as op’-DDD therapy inhibited function of the remaining adrenal gland. Mean length of exogenous steroid replacement therapy for all adenoma pa- tients was 16.8 ± 9.1 months. However, 7 of 13 adenoma patients (53.9%) with greater than 24 months of followup required greater than 24 months of steroid replacement. Six of these 7 patients (85.7%) were never successfully tapered off exogenous steroids (table 4). Regression analysis adjusted for length of followup demonstrated no relationship between preoperative urinary free cortisol levels and length of post- operative steroid replacement requirement (p = 0.96).
Disease recurrence / patient survival. There were no ade- noma recurrences. Ten carcinoma patients (76.9%) had re- currences at a mean of 33 months. There was 1 perioperative death among adenoma patients due to hypoglycemia. Among
carcinoma patients 3 and 5-year survival rates were 41.5 and 31.2%, respectively.
DISCUSSION
We reviewed our large experience with adrenal Cushing’s syndrome to provide further insight into the clinical presen- tation and outcome of this uncommon disorder. In accordance with several other studies, the left adrenal gland and female patients were predominantly affected in our series.2-4,9-11 It is unclear why the left adrenal is preferentially affected. The preponderance of affected female patients may be due to the nature of the presenting symptoms of Cushing’s syndrome including hirsutism and oligomenorrhea, which uniquely af- fect women.
No particular sign or symptom was indicative of adenoma versus carcinoma at presentation. Other groups have noted presence of virilization/hirsutism as a powerful clinical tool in suggesting carcinoma.1,9,12 The theory is that adrenal carcinomas do not convert androgenic steroid precursors to glucocorticoids as efficiently as adenomas resulting in andro- genic steroid excess among carcinoma patients. In our study this was not summarily the case. In fact, while most carci- noma patients presented with mixed endocrine abnormali- ties, several adenoma patients also presented with signs and symptoms of mixed endocrine syndrome. Indeed, among ad- enoma patients in whom data were available 15 of 16 (93.8%) presented with hirsutism (table 1). In addition biochemical evaluation revealed mixed endocrine abnormalities in 6 of 19 (31.6%) adenoma patients. Other investigators have ac- knowledged that adrenal adenomas can produce high levels of androgenic steroids. In their review of 36 patients with Cushing’s syndrome due to adrenal tumors, Bertagna and Orth reported that patients presenting with hirsutism are more likely to suffer from adrenocortical carcinoma than from adenoma.10 Yet in their study 14 of 18 patients (78%) with adrenal adenomas presented with hirsutism. Hence there is clearly a large degree of overlap in presenting signs and symptoms among adenoma and carcinoma patients. In our study only glucose intolerance tended to be more preva- lent among carcinoma patients than adenoma patients (p = 0.13). This prevalence may reflect the trend toward a higher degree of cortisol elevation among carcinoma patients (p = 0.35, table 2).
Similarly no specific hormonal study could distinguish ad- enoma from carcinoma patients. Although there was a trend toward higher 24-hour urinary free cortisol and serum lac- tate dehydrogenase levels among carcinoma patients, these trends were not statistically significant (table 2). Nonethe- less these data probably reflect the larger bulk of biochemi- cally active tumor burden among carcinoma patients. Malunowicz et al in their study of 11 children with adreno- cortical tumors also found that urinary steroid profiling only confirms the presence of an adrenal tumor and can be mis- leading if used to distinguish adrenal adenomas from carci- nomas.13
While not statistically significant, the trend toward higher 17-KS and DHEA-S levels among carcinoma patients in our
| Normal Range | Adenoma | Ca | Adenoma vs. Ca (p value) | |
|---|---|---|---|---|
| 24-Hr. urinary free cortisol | 7-70 Mcg./24 hrs. | 699.9 ± 1379.6 | 1130.6 ± 1058.2 | 0.35 |
| Adrenocorticotrophic hormone | 5-50 Pg./ml. | 6.3 ± 2.4 | 5.8 ± 1.1 | 0.68 |
| 17-OH | 2-6 Mg./24 hrs. | 10.7 ± 7.6 | 23.8 ± 14.4 | 0.12 |
| 17-KS | 4-17 Mg./24 hrs. | 13.0 ± 6.4 | 122.5 ± 190.2 | 0.09 |
| DHEA-S | 0.36-3.2 Mcg/ml. | 1.7 ± 2.5 | 146.4 + 349.2 | 0.35 |
| Testosterone (female pts. only) | 20-70 Ng/dl. | 107.2 ±157.2 | 130.3 ± 82.4 | 0.74 |
| Aldosterone | 2.9-22.5 Ng/dl. | 30.5 ± 40.0 | 44.7 ± 31.7 | 0.51 |
| Lactate dehydrogenase | 100-220 Units/l. | 289.5 ± 84.7 | 575.6 ± 477.5 | 0.06 |
All values are mean plus or minus standard deviation.
| Adenoma (9 pts.) | Ca (8 pts.) | |||||
|---|---|---|---|---|---|---|
| Decreased | Normal | Increased | Decreased | Normal | Increased | |
| Total No. pts. (%) | 0.05 | 0.45 | 5.5 | 0.13 | 0.6 | 3.9 |
| 0.06 | 1.3 | 6.5 | 1.2 | 4.9 | ||
| 0.1 | 1.3 | 2.4 | 15.8 | |||
| 0.16 | 1000 | |||||
| 4 (44.4) | 3 (33.3) | 2 (22.2) | 1 (12.5) | 3 (37.5) | 4 (50.0) | |
| followup | |||
|---|---|---|---|
| Pt. | Length Followup (mos.) | Length Steroid Replacement (mos.) | Comments |
| KW | 83 | 83 | Re-started after at- tempted taper at 14 mos. |
| CF | 31 | 11 | |
| LH | 28 | 28 | Never tapered |
| SJ | 133 | 16 | |
| BY | 105 | 22 | |
| FD | 147 | 147 | Initially tapered at 48 mos., re-started shortly after taper |
| LF | 140 | 13 | |
| DR | 26 | 26 | Never tapered |
| MZ* | 109 | 37 | Successful taper after greater than 24 mos. of steroid replacement |
| SJ | 94 | 94 | Never tapered |
| SS | 170 | 9 | |
| DW | 87 | 87 | Initially tapered at 34 mos., re-started 15 mos. later |
| NR | 253 | 13 | |
* Only 1 patient (MZ) who required greater than 24 months steroid replace- ment was tapered off exogenous steroids successfully.
study warrants further exploration (table 2). For example, 7 of 9 adenoma patients (77.8%) with recorded 17-KS levels presented with normal values, whereas only 3 of 11 (27.3%) carcinoma patients had normal 17-KS levels. In addition, among carcinoma patients with abnormal 17-KS levels, 5 (62.5%) had elevations greater than 3 times the upper limit of normal. Both adenoma patients with abnormally high 17-KS levels presented with only slight elevations.
Likewise 4 of the 9 adenoma patients (44.4%) with re- corded DHEA-S levels actually presented with subnormal values, and the 2 adenoma patients (22.2%) who presented with elevated DHEA-S levels had only mild elevations (table 3). In contrast, only one carcinoma patient (12.5%) presented with a subnormal DHEA-S level, and of the 4 patients (50%) with elevated levels, 2 had values greater than 4 times the upper limit of normal. Yamaji et al reported similar findings among 6 patients with Cushing’s syndrome due to adrenal adenomas.14 Serum DHEA-S levels were significantly lower among patients with Cushing’s syndrome due to adrenal adenomas than were normal controls. They theorized that since ACTH is the major regulator of androgen secretion from the adrenal cortex, decreased ACTH levels found in Cushing’s syndrome would result in decreased DHEA-S lev- els. No comparison, however, was made between adenoma and carcinoma patients with Cushing’s syndrome. Gabrilove et al compared adenoma and carcinoma patients in their study of 8 subjects with adrenal Cushing’s syndrome.15 Mean serum DHEA and DHEA-S levels were below normal range for all adenoma patients and elevated for all carcinoma pa- tients. Although no statistical comparisons were performed, these findings do support the theory that carcinomas may not convert steroid precursors to glucocorticoids as efficiently as adenomas. Hence, while no steadfast rules can be stated with regard to specific preoperative hormonal analysis, it is likely
that patients with adrenal Cushing’s syndrome, normal 17-KS levels and subnormal DHEA-S levels probably have adrenal adenomas, whereas those with greatly elevated 17-KS or DHEA-S levels most likely suffer from adrenocor- tical carcinoma. It is prudent to remember, however, that adenoma patients present with mixed endocrine abnormali- ties, and only tumor size and patient age are reliable preop- erative indicators of adenoma versus carcinoma.
Few studies have examined which radiographic findings distinguish adrenal adenomas from carcinomas. In general, CT is the most useful imaging study, with MRI reserved for indeterminate masses.16,17 Angiography cannot reliably de- lineate adenomas from carcinomas and should be reserved for vascular mapping of large tumors or lesions in which distinguishing the tissue of origin is difficult.17
CT of adrenal adenomas typically demonstrates small masses in which calcifications are uncommon and necrosis is rare.18 Adenomas typically exhibit low attenuation due to high lipid content, and several studies have shown that le- sions with a noncontrast attenuation less than 10 to 15 Houndsfield units should be considered benign.19,20 Adreno- cortical carcinomas are larger with calcifications in approxi- mately 30% of cases. In addition central necrosis is often present.16,18 Our findings appear to agree with those of other authors.
Similarly several studies have examined the use of MRI in distinguishing adrenal adenomas from nonadenomas. Most studies, however, principally consider pheochromocytomas and adrenal metastases and not adrenocortical carcinomas as nonadenomas. Carcinomas typically have increased signal intensity on T2-weighted images. However, as in our study, there is as much as 31% overlap among benign and malig- nant lesions.18,21 Recent studies using fast low-angle shots and magnetic resonance spectroscopic imaging to distinguish adrenal adenomas from carcinomas have been encourag- ing.22,23
In accordance with most other studies, our study demon- strated the mean length of steroid replacement therapy to be 16.8 ± 9.1 months for adenoma patients. Doherty et al re- ported recovery of the hypothalamic-pituitary-adrenal axis at a mean of 15 months with successful withdrawal of steroid replacement therapy at 19 months. All patients were off steroid replacement by 24 months postoperatively.24 How- ever, only 8 patients were included in the study of that group. Bertagna and Orth reported on 58 patients with adrenocor- tical tumors.10 They reported normal response to ACTH stim- ulation test within 12 months postoperatively in 9 of 11 patients for whom data were available. All 11 patients were tapered from steroid replacement therapy by 22 months post- operatively. In our study, however, among the 13 adenoma patients with greater than 24 months of followup, 7 (53.9%) required greater than 24 months of steroid replacement ther- apy. In addition, of these 7 patients, only 1 was successfully tapered off steroids (table 4). While this may seem somewhat surprising, other groups have also demonstrated the need for prolonged exogenous steroid replacement.25-27
Gordon et al reported on 6 patients who required glucocor- ticoid replacement 2.5 to 11 years after adrenalectomy for adrenal Cushing’s syndrome.25 Testing with insulin-induced hypoglycemia and cortisol releasing factor stimulation, they
found that 4 patients (66.7%) still suffered defects in the hypothalamic-pituitary-adrenal axis and continued to re- quire exogenous steroid supplementation. Similarly, in Welbourn’s series of 13 patients with Cushing’s syndrome secondary to adrenal adenomas, 3 patients (23.1%) required lifelong steroid replacement therapy.26 Finally, Välimäki et al reviewed 16 patients who underwent adrenalectomy for Cushing’s syndrome due to adrenal adenomas.27 Of 13 pa- tients (23.1%) with long-term followup 3 demonstrated in- complete recovery of the hypothalamic-pituitary-adrenal axis by ACTH stimulation test. Hence, while most patients with Cushing’s syndrome due to adrenal adenomas can be tapered off steroid replacement therapy within 2 years of adrenalec- tomy, the need for prolonged and perhaps lifelong steroid replacement is real, and probably affects approximately 25% of all patients. In addition, in our study 3 patients who were thought to be successfully tapered off steroid replacement therapy reported reinitiating their use when contacted by telephone (table 4). Previous studies without phone followup may have misclassified these patients as having recovered a normal hypothalamic-pituitary-adrenal axis.
Among adenoma patients in our series there were no re- currences and only 1 disease-specific mortality. This periop- erative death from hypoglycemia emphasizes risks of hypo- cortisolism observed after adrenalectomy. Among carcinoma patients a mean survival of 47.6 ± 56.2 months, and 5-year survival rate of 31.2% are similar to other studies in which mean survival rates of 30 to 40 months and 5-year survival rates of 16 to 25% have been reported.1-4,28 As in these studies, our review also demonstrated that despite a func- tional nature, adrenocortical carcinomas producing Cush- ing’s syndrome are not identified at an earlier stage when compared with patients suffering from nonfunctional adre- nocortical carcinomas. Despite complete surgical extirpation for most patients, and postoperative chemotherapy for all patients, survival remains poor among carcinoma subjects.
CONCLUSIONS
Cushing’s syndrome due to adrenal adenoma or adrenocor- tical carcinoma is rare. Distinguishing adenomas from carci- nomas among these patients can often be challenging. While presenting signs and symptoms and hormonal analysis may suggest benign or malignant disease, there is much overlap between adenoma and carcinoma patients. Adenoma as well as carcinoma patients often present with mixed endocrine syndromes. Only tumor size and patient age are reliable preoperative indicators of adrenal adenoma versus adreno- cortical carcinoma. Surgery is curative for adenoma patients but lifelong steroid replacement may be required. Survival remains poor among carcinoma patients.
Jar-Chi Lee provided statistical analysis.
REFERENCES
1. Pommier, R. F. and Brennan, M. F .: Management of adrenal neoplasms. Curr. Probl. Surg., 27: 657, 1991.
2. Bodie, B., Novick, A. C., Pontes, J. E., Straffon, R. A., Montie, J. E., Babiak, T., Sheeler, L. and Schumacher, P .: The Cleve- land Clinic experience with adrenal cortical carcinoma. J. Urol., 141: 257, 1989.
3. Wooten, M. D. and King, D. K .: Adrenal cortical carcinoma. Epidemiology and treatment with mitotane and a review of the literature. Cancer, 72: 3145, 1993.
4. Zografos, G. C., Driscoll, D. L., Karakousis, C. P. and Huben, R. P .: Adrenal adenocarcinoma: a review of 53 cases. J. Surg. Oncol., 55: 160, 1994.
5. Ross, N. S. and Aron, D. C .: Hormonal evaluation of the patient with an incidentally discovered adrenal mass. New Engl. J. Med., 323: 1401, 1990.
6. Oldfield, E. H., Doppman, J. L., Nieman, L. K., Chrousos, G. P., Miller, D. L., Katz, D. A., Cutler, G. B., Jr. and Loriaux, D. L .: Petrosal sinus sampling with and without corticotropin- releasing hormone for the differential diagnosis of Cushing’s syndrome. New Engl. J. Med., 325: 897, 1991.
7. Nader, S., Hickey, R. C., Sellin, R. V. and Samaan, N. A .: Adre- nal cortical carcinoma: a study of 77 cases. Cancer, 52: 707, 1983.
8. Kaplan, E. L. and Meier, P .: Nonparametric estimation from incomplete observations. J. Amer. Stat. Assoc., 53: 457, 1958.
9. Samuels, M. H. and Loriaux, D. L .: Cushing’s syndrome and the nodular adrenal gland. Endocr. Metab. Clin. N. Amer., 23: 555, 1994.
10. Bertagna, C. and Orth, D. N .: Clinical and laboratory findings and results of therapy in 58 patients with adrenocortical tu- mors admitted to a single medical center (1951 to 1978). Amer. J. Med., 71: 855, 1981.
11. Didolkar, M. S., Bescher, R. A., Elias, E. G. and Moore, R. H .: Natural history of adrenal cortical carcinoma: a clinicopatho- logic study of 42 patients. Cancer, 47: 2153, 1981.
12. Orth, D. N .: Cushing’s syndrome. New Engl. J. Med., 332: 791, 1995.
13. Malunowicz, E. M., Ginalska-Malinowska, M., Romer, T. E., Ruszcynska-Wolska, A. and Dura, M .: Heterogeneity of uri- nary steroid profiles in children with adrenocortical tumors. Horm. Res., 44: 182, 1995.
14. Yamaji, T., Ishibashi, M. Sekihara, H., Itabashi, A. and Yanaihara, T .: Serum dehydroepiandrosterone sulfate in Cushing’s syndrome. J. Clin. Endocr. Metab., 59: 1164, 1984.
15. Gabrilove, J. L., Freiberg, E. K. and Nicolis, G. L .: Peripheral blood steroid levels in Cushing syndrome due to adrenocortical carcinoma or adenoma. Urology, 22: 576, 1983.
16. Dunnick, N. R .: Adrenal carcinoma. Radiol. Clin. N. Amer., 32: 99, 1994.
17. Kolmannskog, F., Kolbenstvedt, A. and Brekke, I. B .: CT and angiography in adrenocortical carcinoma. Acta Radiol., 33: 45, 1992.
18. Dunnick, N. R .: Hanson lecture. Adrenal imaging: current sta- tus. AJR, 154: 927, 1990.
19. Miyake, H., Takaki, H., Matsumoto, S., Yoshida, S., Maeda, T. and Mori, H .: Adrenal nonhyperfunctioning adenoma and non- adenoma: CT attenuation value as discriminative index. Ab- dom. Imaging, 20: 559, 1995.
20. Korobkin, M., Brodeur, F. J., Yutzy, G. G., Francis, I. R., Quint, L. E., Dunnick, N. R. and Kazerooni, E. A .: Differentiation of adrenal adenomas from nonadenomas using CT attenuation values. AJR, 166: 531, 1996.
21. Chang, A., Glazer, H. S., Lee, J. K. T., Ling, D. and Heiken, J. P .: Adrenal gland: MR imaging. Radiology, 163: 123, 1987.
22. Tsushima, Y., Ishizaka, H. and Matsumoto, M .: Adrenal masses: differentiation with chemical shift, fast low-angle shot MR imaging. Radiology, 186: 705, 1993.
23. Leroy-Willig, A., Bittoun, J., Luton, J. P., Louvel, A., Lefevre, J. E., Bonnin, A. and Roucayrol, J. C .: In vivo MR spectroscopic imaging of the adrenal glands: distinction between adenomas and carcinomas larger than 15 mm based on lipid content. A.JR, 153: 771, 1989.
24. Doherty, G. M., Nieman, L. K., Cutler, Jr., G. B., Chrousos, G. P. and Norton, J. A .: Time to recovery of the hypothalamic- pituitary-adrenal axis after curative resection of adrenal tu- mors in patients with Cushing’s syndrome. Surgery, 108: 1085, 1990.
25. Gordon, D., Semple, C. G., Beastall, G. H. and Thompson, J. A .: A study of hypothalamic-pituitary-adrenal suppression follow- ing curative surgery for Cushing’s syndrome due to adrenal adenoma. Acta Endocrinol., 114: 166, 1987.
26. Welbourn, R. B .: Barber’s company symposium on endocrine surgery: current concepts. Brit. J. Surg., 67: 723, 1980.
27. Välimäki, M., Pelkonen, R., Porkka, L., Sivula, A. and Kahri, A .: Long-term results of adrenal surgery in patients with Cushing’s syndrome due to adrenocortical adenoma. Clin. En- docrinol., 20: 229, 1984.
28. Goldfarb, D. A .: Surgical adrenal disorders. Sem. Nephrol., 14: 570, 1994.