Adrenal Surgery: Trends During the Seventies
Bertil Hamberger, MD, Rochester, Minnesota Colin F. Russell, MB, FRCS, Rochester, Minnesota Jonathan A. van Heerden, MB, FRCS (C), FACS, Rochester, Minnesota William H. ReMine, MD, FACS, Rochester, Minnesota Robert C. Northcutt, MD, Rochester, Minnesota Patrick F. Sheedy II, MD, Rochester, Minnesota Anthony J. Edis, MD, FACS, Rochester, Minnesota Duane M. Ilstrup, MS, Rochester, Minnesota
The adrenal gland is a composite endocrine organ with a bipartite embryologic origin. The cortex, which secretes steroid hormones, derives from the mesoderm, and the medulla, which secretes cate- cholamines, originates from the ectodermal neural crest. Hyperplastic and neoplastic conditions af- fecting the adrenal gland may be associated with excess secretion of the various hormones producing a wide variety of clinical syndromes [1,2]. In recent years, preoperative diagnosis and localization of such lesions have improved, and in some of the hyper- plastic conditions, the indications for surgery have changed [1,3].
This study is a review of the surgical experience with adrenal disease at the Mayo Clinic during the period from 1970 through 1979. It examines accuracy and trends in the use of the various adrenal local- ization studies and documents operative approach, postoperative morbidity and mortality, and long- term survival.
Patients and Methods
During the years 1970 through 1979, 315 patients un- derwent surgical exploration with a presumptive diagnosis of adrenal disease. Patients younger than 10 years of age, most of whom had neuroblastomas, were excluded. There were 141 males and 174 females and the median age was 50 years.
Pathology: The pathologic conditions that were en- countered are listed in Table I. Of the 35 (11.1 percent) adrenocortical carcinomas, 17 were cortisol-producing and 2 were aldosterone-secreting. Eleven (14 percent) of the 77 patients with pheochromocytoma were classified as having
From the Departments of Surgery, Internal Medicine, Diagnostic Radiology, and Medical Statistics and Epidemiology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota. Dr. Hamberger is a visiting clinician from the De- partment of Surgery, Karolinska Hospital, Stockholm, Sweden. Dr. Russell is a visiting clinician from the Royal Victoria Hospital, Belfast, Northern Ireland and a recipient of an Ethicon Foundation Grant and a Wellcome Re- search Travel Grant.
Requests for reprints should be addressed to Jonathan A. van Heerden, MD, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905.
multiple endocrine neoplasia, type II. In 10 (3.2 percent) of the patients who underwent surgical exploration for suspected adrenal lesions, no pathology was encountered. The reason for exploration had been either false-positive roentgenographic findings suggesting an adrenal tumor (four patients), or strong clinical or biochemical suspicion of underlying adrenal disease (six patients). The source of the primary tumor in all five patients with adrenal me- tastases was the lung.
The yearly incidence of operations for aldosteronism decreased slightly, whereas the incidence of nonfunctioning tumors increased slightly. Variations in the frequency of other diagnoses showed no clear trend.
Localization procedures: An increasing proportion of adrenal lesions were localized preoperatively during the study period (Figure 1), particularly in the later years when almost all pathologic glands were localized preoperatively. The diagnostic accuracy of the various radiologic modali- ties that were performed is shown in Figure 2. Computer- ized tomography (available during the last half of the decade) had the highest overall diagnostic accuracy (nearly 90 percent). Almost all the false-negative results occurred early in our experience. Toward the end of the study period, computerized tomography was used routinely in the lo- calization of suspected adrenal pathology at the expense of all other modalities, which were being used with a steadily decreasing frequency (Figure 3). In Figure 4 the relation between the diagnostic accuracy of the various methods of localization to the different types of pathologic conditions encountered is shown.
Operation: From 20 to 40 adrenal operations were performed annually throughout the 10 year period. Surgery involved the right adrenal gland in 110 patients and the left adrenal gland in 122 patients. Seventy-seven patients had bilateral operations and six underwent adrenal exploration only. Overall median operating time was 1 hour 55 minutes. A transverse anterior abdominal incision was most com- monly used; however, after 1975 the posterior lumbar ap- proach was used with increasing frequency in patients with small cortical tumors or bilateral hyperplasia. One hundred (32 percent) of the 315 patients required blood transfusions during operation and 15 patients required more than 5 units. Median postoperative hospitalization was 8 days.
| n | % | |
|---|---|---|
| Hypercortisolism | 74 | 24 |
| Bilateral hyperplasia | 50 | |
| Unilateral adenoma | 24 | |
| Hyperaldosteronism | 46 | 15 |
| Bilateral hyperplasia | 7 | |
| Unilateral adenoma | 39 | |
| Adrenocortical carcinoma | 35 | 11 |
| Pheochromocytoma | 77 | 24 |
| Nonfunctioning adenoma | 47 | 15 |
| Miscellaneous | 26 | 8 |
| Virilizing adenoma | 1 | |
| Metastases | 5 | |
| Cysts | 14 | |
| Myelolipoma | 4 | |
| Hematoma | 2 | |
| No pathology | 10 | 3 |
| Total | 315 | 100 |
| METHOD | n | % 20 40 60 | 80 100 |
| Excretory Urogram | 109 | ||
| Nephrotomography | 151 | ||
| Arteriography | 80 | ||
| Radionuclear scan | 64 | ||
| Computerized tomography | 99 |
Figure 2. Diagnostic accuracy of five radiologic methods used to localize adrenal lesions.
Concomitant surgical procedures were performed in a substantial number of patients (37 percent) (Table II). Splenectomy was performed in 53 (27 percent) patients (7 with malignancies) during the course of removing tumors of the left adrenal gland. The pleura was entered in 14 of 51 patients (27 percent) who were operated on by the posterior approach, and only one of these patients required postoperative pleural drainage.
Results
Morbidity and mortality: Postoperative com- plications occurred in 42 (14 percent) of the surviving patients. Postoperative hemorrhage occurred in five patients, four of whom required reexploration. Four patients had clinically significant thromboembolic complications.
Wound infections occurred in 13 patients (4.1 percent); pulmonary complications (pneumonia, significant pleural effusion, and pneumothorax) oc- curred in 14 (4.4 percent); and other significant complications, including gastrointestinal bleeding, abdominal abscesses, septicemia, postoperative renal failure, jaundice, small bowel obstruction, and wound dehiscence, occurred in 14 (4.4).
100
%
80
60
40
20
0
1970
‘71
‘72
‘73
‘74
‘75
‘76
‘77
‘78
‘79
Years
100
%
Computerized tomography
80
· · Arteriography
- Radionuclear scan
"" Nephrotomography
60
40
20
0
1970
‘71
‘72
‘73
‘74
‘75
‘76
‘77
78
‘79
Years
The overall operative mortality rate was 3.5 per- cent (11 patients). All deaths occurred after operation for Cushing’s syndrome, pheochromocytoma, or malignancy (Table III).
Survival: Survival after adrenal surgery (Figure 5) was calculated according to the method of Kaplan and Meier [4] excluding operative mortality. Ex- pected survival for healthy persons of the same age and sex was calculated from a life table of the upper Midwest states. In patients with hypercortisolism due to bilateral cortical hyperplasia, the 5 year sur- vival probability was significantly less than that in matched control subjects (76 versus 98 per- cent, p ≤0.01). The 2 year survival probability in patients with adrenal carcinoma was 34 percent. There was no significant decrease in the 5 year sur- vival probability in patients with functioning or nonfunctioning adenoma or in patients with pheo- chromocytoma.
Comments
A wide variety of pathologic conditions of the adrenal glands was encountered in this review. The numerical distribution of patients according to type-specific adrenal disease has not changed much over the last decade. We were surprised that despite the resurgence of interest in transsphenoidal pitu-
☒ Arteriography
☐ Nephrotomography
☒ Radionuclear scan
☒ Computerized tomography
100
%
75
50
25
0
Hyperplasia
Adenoma
Aldosteronism
Adrenal carcinoma
Pheochromocytoma
Non-functioning adenoma
Hypercortisolism
| Diagnosis | Procedure (n) | ||
|---|---|---|---|
| Splenec- tomy | Cholecy- stectomy | Other | |
| Hypercortisolism | |||
| Bilateral hyperplasia | 6 | 4 | 1 |
| Unilateral adenoma | 2 | 1 | 1 |
| Aldosteronism | 8 | 1 | 9 |
| Adrenocortical carcinoma | 7 | 1 | 10 |
| Pheochromocytoma | 17 | 8 | 23 |
| Nonfunctioning adenoma | 6 | 5 | 6 |
| Other | 4 | 3 | 8 |
| No pathology | 3 | 0 | 1 |
| Total | 53 | 23 | 59 |
itary surgery for Cushing’s disease [1,5], the number of patients with this entity did not decrease notice- ably. The number of patients with primary aldoste- ronism decreased slightly during the latter part of the study period because patients with cortical hyper- plasia are no longer treated surgically [1,3]. In con- trast, our opinion continues to be that catechol- amine-producing medullary hyperplasia should be treated by total adrenalectomy to avoid recurrence and potential malignancy [6].
Although there is little disagreement that patients with functioning adrenal neoplasms should be treated surgically, opinion about nonfunctioning tumors is not unanimous since less is known about their natural history. The number of such tumors that have been removed surgically has increased in the past 3 to 4 years, mainly because of serendipitous detection by computerized tomography. Comput- erized tomography has similarly contributed to the incidental detection of cysts, myelolipomas, and hematomas [1,7-9]. Whether or not small, non- functioning tumors should be surgically removed is
| Diagnosis | Cause |
|---|---|
| Hypercortisolism | Acute pancreatitis Pulmonary embolism Pulmonary and cardiac failure Metastatic medullary thyroid carcinoma |
| Hypercortisolism Hypercortisolism Hypercortisolism | |
| Adrenocortical carcinoma Adrenocortical carcinoma Adrenocortical carcinoma Pheochromocytoma | Hemorrhage Pulmonary embolism Metastatic disease Hemorrhage Aspiration pneumonia Cardiovascular insufficiency Cerebrovascular accident |
| Pheochromocytoma Pheochromocytoma Pheochromocytoma |
100
%
(n=73)
90
Pheochromocytoma
(n=23)
Hypercortisolism, adenoma
80
70
(n=47)
Hypercortisolism, hyperplasia
60
50
40
30
(n=39)
20
Primary carcinoma
10
0
0
1
2
3
4
5
Years
unclear. In practice, we are currently observing a number of patients whom we have elected to receive nonsurgical treatment.
In 1970, approximately 50 percent of adrenal dis- ease was localized preoperatively and toward the end of the study period almost 100 percent. The adrenal iodocholesterol scan, in particular, proved to be highly accurate in localizing functional adrenocortical disease [10]. However, the major advance in localizing adrenal disease was, without doubt, the introduction in late 1975 of computerized tomography, which has proved to be highly accurate for all mass lesions, in particular those of the adrenal gland. The success of computerized tomography has resulted in greatly diminished roles for both arteriography and ne- phrotomography [2,7,9].
A transverse (unilateral or bilateral) epigastric incision was used in most patients, with excellent exposure. On rare occasions-for large, malignant tumors-thoracoabdominal exposure was utilized. The posterior lumbar extraperitoneal approach is currently favored for removal of small cortical tumors and hyperplasia, unless additional, associated in- traabdominal disease that requires surgical attention is present. Postoperative morbidity and mortality and hospital stay appear to be decreased with this approach when compared with the anterior trans- abdominal approach [11].
Surgical morbidity in this series compared favor- ably with that in other published series [12-14]. The low incidence of thromboembolic complications after surgery in patients with hypercortisolism is partic- ularly noteworthy. Overall operative mortality for our series was 3.5 percent. All deaths occurred in high- risk patients with malignant adrenal tumors, severe hypercortisolism, or catecholamine excess. Death was caused by uncontrollable intraoperative hemorrhage or cardiogenic shock in 3 of the 11 patients who died.
The survival probabilities after adrenalectomy for patients with cortical adenoma or pheochromocy- toma were not significantly reduced. However, in patients with Cushing’s disease due to hyperplasia, the 5 year survival probability was significantly de- creased to 76 percent (control probability 98 per- cent). This decrease probably resulted from both the adrenocorticotropic-hormone-producing lesion in the pituitary gland and the problems that are encoun- tered with steroid supplementation. Primary adrenal carcinoma has a poor prognosis, with a 2 year survival probability of only 34 percent.
This study of the past decade indicates that adrenal operations can be performed with low mor- bidity and operative mortality. Localization of adrenal lesions by computerized tomography is the most important recent advance in the management of adrenal disease. With continued refinements in this technology, localization by this noninvasive method should continue to be highly accurate, thus
aiding and simplifying the sometimes complex and prolonged work-up in this interesting and challenging group of patients.
Summary
Surgical experience with adrenal disease from 1970 to 1979 was reviewed in 315 patients. The pathologic conditions that were encountered were hypercor- tisolism (74 patients), hyperaldosteronism (46 pa- tients), adrenocortical carcinoma (35 patients), pheochromocytoma (77 patients), and nonfunc- tioning adenoma (47 patients). In addition, 5 patients with metastatic lesions, 14 with cysts, and 4 with myelolipoma were surgically treated. The accuracy of localizing adrenal lesions increased from about 50 percent to almost 100 percent during the decade studied. The increase was due mainly to the intro- duction of computerized tomography, the most im- portant advance in the management of adrenal dis- ease. The present study shows that adrenal surgery can be performed with low morbidity and mortality. Operative deaths were confined to patients with malignant disease or increased secretion of cortisol or catecholamines. Only patients with adrenocortical carcinoma (2 year survival probability, 34 percent) or hypercortisolism due to cortical hyperplasia (5 year survival probability, 76 percent) had signifi- cantly decreased survival.
References
1. Edis AJ, Ayala LA, Egdahl RH. Manual of endocrine surgery. New York: Springer-Verlag, 1975:123-208.
2. Brennan MF. The adrenal gland. In: Hellman S. Rosenberg S, eds. Principles and practice of oncology. Philadelphia: JB Lippincott, 1982:985-1000.
3. Hunt TK. Schambelan M, Biglieri EG. Selection of patients and operative approach in primary aldosteronism. Ann Surg 1975;182:353-60.
4. Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Statis Assoc 1958;53:457-81.
5. Javadpour N, Woltering EA, Brennan MF. Adrenal neoplasms. Curr Probl Surg 1980;17:1-52.
6. Sizemore GW, Heath H III, Carney JA. Multiple endocrine ne- oplasia type 2. Clin Endocrinol Metab 1980;9:299-315.
7. Hattery RR, Sheedy PF, Stephens DH, van Heerden JA. Com- puted tomography of the adrenal gland. Semin Roentgenol 1981;16:290-300.
8. Budd DC, Fink DL. Cysts of the adrenal gland. Am Surg 1979;45:649-52.
9. Reynes CJ, Churchill R, Moncada R, Love L. Computed to- mography of adrenal glands. Radiol Clin North Am 1979; 17:91-104.
10. Thrall JH, Freitas JE, Beierwaltes WH. Adrenal scintigraphy. Semin Nucl Med 1978;8:23-41.
11. Russell CF, Hamberger B, van Heerden JA, Edis AJ, Ilstrup DM. Adrenalectomy: anterior or posterior approach? Am J Surg (in press).
12. James VHT. The adrenal gland. New York: Raven Press, 1979:165-95.
13. Kümmerle F, Lenner V, Cordes U, Günther R. Nebennieren- chirurgie: bericht über 154 operative Eingriffe. Dtsch Med Wochenschr 1980; 105:679-85.
14. Scott HW Jr, Liddle GW, Mulherin JL Jr, Mckenna TJ, Stroup SL, Rhamy RK. Surgical experience with Cushing’s disease. Ann Surg 1977;185:524-33.