SURGERY FOR ADRENAL TUMOURS: IS OPERATION FOR THE SMALL INCIDENTAL TUMOUR APPROPRIATE?
DOUGLAS E. WOOD, LEIGH DELBRIDGE AND THOMAS S. REEVE Department of Surgery, The University of Sydney at Royal North Shore Hospital, St Leonards, New South Wales
The last 15 years have seen a change in the indications for adrenalectomy. On the one hand there has been a decline in the number of operations for metastatic breast cancer due principally to the introduction of effective medical therapy. Moreover, the introduction of trans-sphenoidal techniques has reduced the number of adrena- lectomies for Cushing’s Disease. In contrast the introduction of CT scanning has seen an increase in the diagnosis of incidental adrenal masses, most of which turn out to be benign non-functioning adrenal cortical adenomas. During the period 1970-86, 124 adrenalectomies were performed at Royal North Shore Hospital. From 1970 to 1976 the primary indication for adrenalectomy was metastatic breast carcinoma (84%). Over the past three years one of the primary indications has been the discovery of an asymptomatic, non-functional adrenal mass on CT scan (26%) and, of these tumours, 95% have been less than 5 cm in size. Yet no incidentally discovered adrenal mass was an adrenocortical carcinoma and no adrenocortical carcinoma was smaller than 7 cm in diameter. We recommend that an asymptomatic adrenal mass less than 6 cm in size, with no evidence of function, should be managed conservatively.
Key words: adrenal carcinoma, adrenal tumours, CT scanning.
Introduction
Since the early 1970s the indications for adrena- lectomy have been changing with the advent of medical therapy for stage IV breast cancer and improvements in hypophysectomy for Cushing’s Disease. Whilst adrenalectomy for metastatic breast cancer has been generally eliminated, there has been an increase in the diagnosis of asympto- matic, non-functioning adrenal masses with the widespread introduction and increasing sophistica- tion of CT scanning.
There is increasing evidence that many of these asymptomatic, non-functioning adrenal tumours which are discovered incidentally (‘incidentalo- mas’) do not require surgical removal.1-6 Endo- crine function, CT characteristics, needle biopsy and size have all been proposed as parameters for determining the significance of adrenal masses.
The aim of this study was to evaluate our experi- ence at Royal North Shore Hospital with adrenal surgery and to determine guidelines for the investi- gation and management of adrenal tumours.
Correspondence: Professor T. S. Reeve, Department of Sur- gery, Wallace Freeborn Block, Royal North Shore Hospital, St Leonards, NSW 2065, Australia.
Accepted for publication 16 April 1987.
Patients and methods
A retrospective study was undertaken of all patients undergoing adrenalectomy from January 1970 to December 1986. Information on all patients was obtained from review of the hospital medical records and private notes.
Adrenalectomies secondary to other procedures such as nephrectomy and retroperitoneal lymph node dissection were excluded, leaving a total of 124 patients for review. Of these, five records were incomplete or unobtainable and were excluded. There were thus 119 patients included in this study with a mean age of 51.5 years (range 28-80).
Information obtained included indications for surgery, procedure performed, size and pathologi- cal diagnosis of the specimens, and the complica- tions of the operation.
Results
The incidence of adrenalectomy over the study period is shown in Fig. 1. Sixty-eight women un- derwent bilateral adrenalectomy for stage IV breast carcinoma. Of these, 39% showed metastatic breast cancer in one or both adrenal glands. Metas- tatic breast carcinoma was the major indication for adrenalectomy from 1970 to 1976, rapidly declining after this period with the advent of medical adrenalectomy and tamoxifen therapy. No
20
breast cancer
other
10
0
70
75
80
85
adrenalectomies for breast cancer have been per- formed since 1980.
Excluding those patients where the indication was breast cancer, there were 51 adrenalectomies performed for indications listed in Table 1. Func- tional tumours (n=34) were defined as adrenal masses with a documented abnormality of endo- crine secretion; symptomatic masses (n = 5), as having specific local symptoms as an indication for imaging in the region; and incidental masses (n=12). as being discovered during investigation for systemic or non-specific complaints. In this group of patients the mean age was 50.3 years (range 28-78) and the sex distribution was (F:M) 1.68:1.
Of those patients with functional tumours, Cush- ing’s Syndrome was the most common. Eight of these had Cushing’s Disease, five undergoing bilateral adrenalectomy early in the series whilst the three patients operated on for Cushing’s Disease after 1975 had recurrent or persistent disease after transphenoidal hypophysectomy. Five patients with Cushing’s Syndrome had functioning cortical ad- enomas, two had nodular hyperplasia and the remaining five had functioning carcinomas. Conn’s Syndrome occurred in six patients and all of these had functioning adenomas. Eight patients had phaeochromocytomas and of these only one has since developed metastatic disease.
Five patients presented with symptomatic masses and those symptoms and the pathology found are
| Metastatic breast carcinoma Functioning tumours | 68 |
| Cushing's Syndrome | 20 |
| Phaeochromocytoma | 8 |
| Conn's Syndrome | 6 |
| Symptomatic masses | 5 |
| Incidental masses | 12 |
| Symptoms | Pathology |
|---|---|
| 1. RUQ pain, hypertension | 3.3 cm adenoma |
| 2. Weight loss, abdominal mass | 19 cm carcinoma |
| 3. IVC obstruction | 10 cm carcinoma |
| 4. LUQ pain | 5.5 cm phaeochromocytoma |
| 5. LUQ pain, abdominal | 15 cm cyst |
| mass |
listed in Table 2. The most common symptom was upper quadrant or flank pain.
Table 3 documents the indications for imaging, and subsequent pathology, in those 12 patients who were shown to have an incidental adrenal tumour. This group excludes those patients whose subse- quent investigation showed evidence of endocrine function. Retrospectively, the only incidental adre- nal mass that warranted adrenalectomy in this group was the large 8 cm neuroblastoma. There were no incidentally discovered adrenocortical car- cinomas in this group.
Figure 2 compares the size of the tumour (removed specimen) for the adrenocortical adeno- mas and carcinomas. It can be seen that there was only one adenoma measuring greater than 6 cm (7×4×6 cm adenoma) whilst there were no carci- nomas less than 6 cm.
There was only one death in the group (excluding patients operated on for stage IV breast cancer). This occurred in a 78 year old patient with dissemi- nated non-Hodgkin’s lymphoma. Postoperative complications occurred in 30% of these cases. There were six wound infections, four patients with pneumonia/consolidation, three with postoperative
| Indications for imaging | Pathology | |
|---|---|---|
| 1. | Nausea, vomiting | 2.8 cm adenoma |
| 2. | Weight loss, anorexia | 3.7 cm nodular hyperplasia |
| 3. | Retroperitoneal | adrenal infarct |
| haematoma | ||
| 4 | Anaemia | lymphoma |
| 5 | Carcinoma follow-up | 6 cm metastasis |
| 6. | Hypertension | 7 cm adenoma |
| 7. | Diffuse abdominal pain | 3 cm adenoma |
| 8. | Anaemia, weight loss | 8 cm metastasis |
| 9. | Headaches | 8 cm neuroblastoma |
| 10. | Hypertension, fatigue | leiomyoma |
| 11. | Fatigue | 2 cm adenoma |
| 12. | Abdominal pain | 3.5 cm adenoma |
10
benign adrenal adenoma
adrenocortical carcinoma
5
0
1
3
5
7
9
11
13
>15
size (cm)
haemorrhage, two with pneumothorax and four other complications.
Discussion
The widespread availability of increasingly sophis- ticated CT scanning has led to the problem of how to deal with the incidentally discovered adrenal tumour. Autopsy series have shown that benign adrenocortical adenomas occur in 1.4 to 8.7% of the population7-10 whilst radiological series have reported adrenal masses in 0.6-1.3% of patients undergoing upper abdominal CT scanning. 2.3,5,11 This raises the question of when to operate on asymptomatic, non-functioning adrenal masses. The concern lies with recognising and removing the adrenocortical carcinoma, whilst saving the patient with the clinically insignificant adenoma the mor- bidity and expense of an unnecessary operation.
Several criteria have been proposed for determin- ing the nature of an adrenal mass, namely endo- crine function, CT characteristics, fine needle biopsy and size.
With regard to endocrine function, between 4 and 76% of adrenocortical carcinomas are ‘non- functional’ depending on the degree of diligence of the investigation.1 Since the enzymatic machinery of carcinomas may well be inefficient, 17-OH- corticosteroids, 17-ketosteroids and 17-ketogenic steroids should be measured even in the absence of clinical Cushing’s Syndrome.1 A clinically un- suspected phaeochromocytoma should be excluded by urinary VMA, metanephrines and catechola- mines and hypertensive patients should be investi- gated for mineralocorticoid excess.12 Oestrogen and testosterone levels need only be obtained if clinically indicated. In this study, 71% of the ad- renocortical carcinomas had documented endocrine hyperfunction.
The characteristics of the tumour on CT scanning may be useful with benign adenomas generally hav- ing a round or oval, well circumscribed appearance
whilst carcinomas may have an irregular contour with invasion of surrounding structures.6 Fine needle aspiration may be useful in diagnosing meta- static adrenal malignancy13 or benign cysts, 14 but is not generally helpful in differentiating benign from malignant adrenocortical lesions.
Many authors have suggested that size is the best criterion for deciding which asymptomatic, non- functional adrenal masses require adrenalectomy, 1-3,11,15 with most recommending exploration for tumours greater than 5-6 cm. Whilst benign adre- nal adenomas are quite common, adrenocortical carcinomas are rare, with a reported annual inci- dence of between 1 in 600 000 to 1 in 1.6 million.’ Since an endocrine abnormality can be detected in around two-thirds of carcinomas, the incidence of endocrinologically silent carcinoma is exceedingly small.
In our series there were seven adrenocortical car- cinomas, five with Cushing’s Syndrome and the remaining two with local symptoms and signs. The smallest carcinoma was 7 cm with a mean of 12.3 cm, a finding comparable to most other series. In the Mayo Clinic series, only one of 62 carcinomas was less than 6 cm (5 cm) whilst in the Roswell Park experience, only one in 42 carcinomas was less than 5 cm and was picked up incidentally in a patient dying of leukaemia.16,17 No ‘inciden- talomas’ in our series turned out to be carcinomas. There were three other solid adrenal masses greater than 6 cm, including a neuroblastoma, a metastatic carcinoma and an adenoma.
In a review of four autopsy series, there were only three adenomas greater than 6 cm in 12 000 autopsies.7-10 Copeland has extrapolated these data and has concluded that even with masses greater than 6 cm, 60 adrenalectomies would have to be done on biochemically silent adenomas to remove one carcinoma, whilst adrenalectomy for smaller masses would remove carcinomas much less often.1 The more aggressive approach of some other authors,2,3,11 exploring all incidental adrenal tumours greater than 3-4 cm, will remove a large number of benign adenomas unnecessarily.
A complicating factor is the presence of meta- static malignancy in the adrenals. In one review, 20% of patients had an adrenal presentation of an unknown primary malignancy and 73% of patients with known malignancy had unsuspected adrenal disease.2 In our series, two of 12 ‘incidentalomas’ were secondary to metastatic disease. Thus a care- ful search for a primary malignancy should be made, and a fine needle biopsy may differentiate adrenal from non-adrenal pathology.
The high incidence of adrenal ‘incidentalomas’ means that surgical removal of all such lesions cannot be warranted just to exclude the remote possibility of malignancy. We would therefore
recommend the following approach to the adrenal mass: patients with an adrenal tumour that have local symptoms or signs, or with clinical or bio- chemical evidence of endocrine hyperfunction, warrant an adrenalectomy because of the higher incidence of malignancy. Patients with asympto- matic, non-functioning masses need to be carefully screened for primary malignancy, especially lung, breast, colon and melanoma. Fine needle aspiration biopsy may be helpful to differentiate metastatic malignancy. In addition. the aspiration of clear flu- id from a cystic lesion confirms its benign nature.
For the remaining masses, those smaller than 6 cm should be followed with sequential CT scans at 2. 6 and 18 months. operating if the mass enlarges and stopping if it remains stable in size over that period.1 The risk of carcinoma in masses greater than 6 cm seems sufficient to tip the balance in favour of adrenalectomy although one could argue. even then, that the morbidity and expense still out- weigh the potential benefit to the patient.
If these guidelines had been applied to the group of patients reported in this series, no case of adre- nocortical carcinoma would have been missed, and only one unnecessary adrenalectomy for a benign adrenal adenoma would have been performed.
Perhaps improved imaging techniques will fur- ther refine the differentiation as there are promising early reports that magnetic resonance imaging can distinguish between benign and malignant adrenal ‘incidentalomas *. 18.19
Acknowledgments
This study was only made possible with the assis- tance of members of the sections of Endocrinology, General Medicine and Renal Medicine, who were responsible for the initial investigation and manage- ment of many of the patients in this series. The study was supported by the Surgical Research Trust Fund. Royal North Shore Hospital, and the Con- solidated Medical Research Fund, University of Sydney.
References
1. COPELAND P. M. (1984) The incidentally discovered adrenal mass. Ann. Surg. 199, 116-22.
2. BEI.LDEGRUN A .. HUSSAIN S., SELTZER S. E., LOUGHLIN K. R., GITTES R. F. & RICHIE J. P. (1986) Inciden- tally discovered mass of the adrenal gland. Surg. Gyn. Obs. 163. 203-8.
3. PRINZ R. A .. BROOKS M. H .. CHURCHILL R. et al.
(1982) Incidental asymptomatic adrenal masses detected by computed tomographic scanning: is operation required? J.A.M.A. 248, 701-4.
4. GEEI.HOED G. W. & DRUY E. M. (1982) Management of the adrenal “incidentaloma”. Surgery 92, 866-74.
5. ABECASSIS M .. MCLOUGHLIN M. J., LANGER B. & KUDLOW J. E. (1985) Serendipitous adrenal masses: prevalence, significance and management. Am. J. Surg. 149. 783-8.
6. MITNICK J. S., BOSNIAK M. A., MEGIBOW A. J. & NAIDICH D. P. (1983) Non-functioning adrenal ad- enomas discovered incidentally on computed tomog- raphy. Radiology 148, 495-9.
7. SHAMMA A. H., GODDARD J. W. & SOMMERS S. C. (1958) A study of the adrenal status in hypertension. J. Chronic. Dis. 8. 587-95.
B. Russi S .. BLUMENTHAL H. T. & GRAY S. H. (1945) Small adenomas of the adrenal cortex in hypertension and diabetes. Arch. Int. Med. 76, 284-9.
9. KOKKO J. P., BROWN T. C. & BERMAN M. M. (1967) Adrenal adenoma and hypertension. Lancet 1, 468- 70.
10. HEDELAND H., OSTBERG G. & HOKFELT B. (1968) On the prevalence of adrenocortical adenomas in an autopsy material in relation to hypertension and dia- betes. Acta Med. Scand. 184, 211-4.
11. GLAZER H. S., WEYMAN, P. J., SAGEL S. S., LEVITT R.G. & McLENNAN B. L. (1982) Nonfunctioning adrenal masses: incidental discovery on computed tomography. Am. J. Roentgenol. 139, 81-5.
12. STOKES G. S., CAMPBELL R., KARPLUS T. E. & MCCARTHY S. W. (1984) Surgical management of primary aldosteronism (Conn’s syndrome), a correct- able cause of hypertension. Aust. N.Z. J. Surg. 54, 451-6.
13. BERKMAN W. A., BERNADINO M. E., SEWELL, C. W., PRICE R. B. & SONES P. J. (1984) The computed tomography-guided adrenal biopsy. Cancer 53, 2098-103.
14. SCHIEBLE W., COEL M., STEMERS P. T. & SIEGEL H. (1977) Percutaneous aspiration of adrenal cysts. Am. J. Roentgenol. 128, 1013-6.
15. O’LEARY T. J. & OOI T. C. (1986) The adrenal in- cidentaloma. Can. J. Surg. 29, 6-8.
16. HENLEY D. J., VAN HEERDEN J. A., GRANT C. S., CARNEY J. A. & CARPENTER P. C. (1983) Adrenal cortical carcinoma - a continuing challenge. Sur- gery 94, 926-31.
17. DIDOLKAR M. S., BESCHER R. A., ELIAS E. G. & MOORE R. H. (1981) Natural history of adrenal cor- tical carcinoma: a clinicopathological study of 42 patients. Cancer 47, 2153-61.
18. GLAZER G. M., WOOLSEY E. J., BORRELLO J. et al. (1986) Adrenal tissue characterization using MR imaging. Radiology 158, 71-9.
19. REINIG J. W., DOPPMAN J. L., DWYER A. J., JOHNSON A. R. & KNOP R. H. (1986) Adrenal masses differen- tiated by MR. Radiology 158, 81-4.