Adrenal surgery for hypercortisolism- surgical aspects
Jon A. van Heerden, MB, FRCS(C), FACS, William F. Young, Jr., MD, Clive S. Grant, MD, FACS, and Paul C. Carpenter, MD, Rochester, Minn.
Background. Patients with endogenous hypercortisolism are thought to be at high risk for adrenalectomy and may experience significant postoperative surgical mortality/morbidity. Methods. From 1981 through 1991, 91 patients underwent adrenal resection for endogenous hypercortisolism. Causes were adrenal-dependent Cushing’s syndrome (50%), pituitary-dependent Cushing’s syndrome (27%), and an ectopic adrenocorticotropic hormone-secreting tumor (23%). Causes of adrenal-dependent Cushing’s syndrome were adrenocortical adenoma (72%), bilateral nodular hyperplasia (20%), and adrenocortical carcinoma (8%). Comparative mean length of hospitalization for patients undergoing unilateral anterior versus posterior approach was 8 versus 6 days, and bilateral anterior versus posterior was 11 versus 6 days.
Results. Operative mortality was 2.6%. Only one patient had a wound infection, and no patient had either a venous thrombosis or a pulmonary embolism. Delayed wound healing occurred in three patients.
Conclusions. (1) Adrenal surgery can be performed today with low morbidity/mortality. (2) Although there is an effect of hypercortisolism on wound healing, infection, diabetes, hypertension, coronary artery disease, and pulmonary embolism, it was possible to perform adrenalectomy surgically with acceptable morbidity and mortality. (3) These results may serve as a standard against which laparoscopic adrenalectomy may be compared. (SURGERY 1995;117:466-72.)
From the Departments of Gastroenterologic and General Surgery, the Division of Hypertension and Internal Medicine, and the Division of Endocrinology/Metabolism and Internal Medicine, Mayo Clinic and Mayo Foundation, Rochester, Minn.
ADRENAL RESECTION IS INDICATED for a wide variety of pathologic conditions affecting the adrenal cortex. When the cortex becomes hyperfunctional (endogenous hypercortisolism), the primary problem may originate in the adrenal cortex per se, whereas in other circum- stances the adrenal cortex is secondarily affected because of disease at a distant site. Patients with hypercortiso- lism, whether the cause is endogenous or exogenous, have traditionally been regarded as at high risk when undergoing any surgical procedure. The principal risks have been those related to decreased wound healing capabilities (decreased fibroblastic activity) and an in- creased risk of infection (decreased immunologic com- petence). To evaluate these risks and to assess the cur- rent indications for and surgical results of adrenal resection in patients with endogenous hypercortisolism, the recent experience with a cohort of patients undergo-
Accepted for publication Sept. 29, 1994.
Reprint requests: Jon A. van Heerden, MD, Mayo Clinic, 200 First St. SW, Rochester, MN 55905.
Copyright @ 1995 by Mosby-Year Book, Inc. 0039-6060/95/$3.00 + 0 11/56/61262
ing adrenal resection for hypercortisolism at a single in- stitution was retrospectively reviewed.
DIAGNOSTIC ASPECTS
Cushing’s syndrome is caused by hypersecretion of corticotropin (adrenocorticotropic hormone [ACTH]) (ACTH-dependent Cushing’s syndrome) or by primary adrenal hypersecretion of glucocorticoids (ACTH-in- dependent Cushing’s syndrome) (Table I). The diag- nostic evaluation of Cushing’s syndrome typically pro- ceeds as outlined in Table II. First, hypercortisolism must be suspected on the basis of the known manifesta- tions of the disorder and then confirmed with measure- ments of serum and 24-hour urine cortisol concentra- tions. Autonomous hypercortisolism (Cushing’s syn- drome) is confirmed with the low-dose dexamethasone suppression test (dexamethasone 0.5 mg orally every 6 hours for 48 hours); a 24-hour urinary cortisol excretion of 20 µg or more with low-dose dexamethasone admin- istration establishes the diagnosis. The low-dose dexa- methasone suppression test is not necessary in patients with obvious stigmata on physical examination and baseline 24-hour urinary cortisol measurements in-
Table I. Causes of Cushing’s syndrome
ACTH-dependent Pituitary Corticotropic adenoma Corticotropic multinodular hyperplasia Ectopic
Ectopic ACTH-secreting tumor Ectopic CRH-secreting tumor Ectopic ACTH- and CRH-secreting tumor
ACTH-independent Unilateral adrenal disease Adenoma Carcinoma
Bilateral adrenal disease Macronodular hyperplasia PPNAD
creased threefold or more above the upper limit of nor- mal. Then the plasma ACTH concentration classifies the subtype of hypercortisolism as ACTH-dependent (normal to high levels of ACTH) or ACTH-indepen- dent (undetectable suppressed ACTH).
In patients with ACTH-dependent hypercortisolism, the high-dose dexamethasone suppression test (dexa- methasone 2 mg orally every 6 hours for 48 hours) helps differentiate pituitary from ectopic ACTH hypersecre- tion; a 24-hour urinary cortisol excretion that decreases 50% or more from baseline values is consistent with pi- tuitary-dependent disease. The overall accuracy with more than 50% suppression is 80%; if more than 90% suppression is found, the accuracy of this test increases to 86%.1 The overnight 8 mg dexamethasone suppres- sion test may be substituted for the 2-day high-dose study. Dexamethasone is administered at 11 PM in a single 8 mg dose. Pl. na for cortisol measurement is drawn the next morning at 8 AM. The criterion for pos- itive response (suppressibility) is a plasma cortisol level 50% or less than the baseline level, a finding consistent with pituitary disease (sensitivity, 89%; specificity, 100%).2 However, some types of ectopic ACTH-secret- ing tumors (e.g., bronchial carcinoid) may also suppress with high-dose dexamethasone administration.3 When a pituitary tumor is not found on computerized pituitary imaging, further evaluation with computerized chest- upper abdominal imaging and inferior petrosal sinus sampling with ovine corticotropin-releasing hormone (oCRH) stimulation4 are indicated. A peak inferior petrosal sinus: peripheral vein ACTH ratio after oCRH administration of 3.0 pg/ml or more has a 100% sensi- tivity and 100% specificity in determining the type of ACTH-dependent Cushing’s syndrome. 4
In patients with ACTH-independent hypercortiso- lism, the high-dose dexamethasone suppression test
Table II. Diagnostic evaluation of Cushing’s syndrome
1. Confirmation of hypercortisolism Clinical assessment
Baseline glucocorticoids (serum cortisol levels, urinary- free cortisol excretion) and low-dose dexamethasone suppression test
2. Subtype diagnosis Plasma ACTH High-dose dexamethasone suppression test
3. Localization studies Directed computerized imaging Inferior petrosal sinus sampling for ACTH with oCRH stimulation
shows no suppression in urinary cortisol excretion. In these patients computerized imaging of the adrenal glands usually indicates the type of adrenal disease (Table I).
MATERIAL AND METHODS
During the 11-year period from 1981 through 1991, 91 patients underwent adrenal resection for endogenous hypercortisolism at our institution. The causes of the hypercortisolism were varied (Figure).
Adrenal-dependent Cushing’s syndrome. Forty- six patients (51%) underwent adrenal resection for ad- renal-dependent Cushing’s syndrome. In this group 93% (43 patients) were female (mean age, 46 years), and three patients were male (mean age, 48 years). The causes of Cushing’s syndrome were a benign adreno- cortical adenoma in 33 patients (72%), an adrenocorti- cal carcinoma in four patients (9%), and bilateral nod- ular adrenal hyperplasia in nine patients (20%). The nodular hyperplasia was either macronodular (six pa- tients) or micronodular (primary pigmented nodular adrenal disease [PPNAD]) (three patients).
Cushing’s disease. Twenty-four patients underwent bilateral total adrenalectomy for Cushing’s disease af- ter unsuccessful transsphenoidal exploration. These 24 patients had collectively undergone a total of 29 trans- sphenoidal operations, and hypercortisolism had per- sisted after operation in all patients. Of these patients 18 (75%) were female (mean age, 38 years) and six (25%) were male (mean age, 45 years). These 24 patients rep- resented 11% of all patients undergoing transsphenoidal surgery for Cushing’s disease during the study period.
Ectopic ACTH-producing tumors. Twenty-one patients underwent bilateral total adrenalectomy for hypercortisolism caused by an ectopic ACTH-produc- ing tumor. Thirteen of these patients (62%) were female and eight were male, with a mean age of 52.0 and 44.0 years, respectively. It is of note that five (24%) of these
Ectopic ACTH (n=21)
23%
50%
Adrenal-dependent Cushing’s Syndrome (n=46)
27 %
Cushing’s Disease (n=24)
| Source | n |
|---|---|
| Unknown | 9 |
| Pancreatic islet cell tumor | 4 |
| Bronchial carcinoid tumor | 2 |
| Thymic carcinoid tumor | 2 |
| Medullary thyroid carcinoma | 2 |
| Pheochromocytoma | 1 |
| Prostatic carcinoma | 1 |
21 patients had undergone a prior unsuccessful, misdi- rected transsphenoidal hypophysectomy (one patient underwent two pituitary operations) before adrenal re- section with the preoperative diagnosis of Cushing’s disease. The sites of the ectopic ACTH-producing tu- mors are depicted in Table III.
Eleven (52%) of the patients in this subgroup died at a mean interval of 18 months after operation. The cause of death was unknown in one patient, and two patients died of pneumonia and cardiac arrhythmia, respec- tively. The remaining eight patients died of metastatic disease caused by pancreatic islet cell carcinoma (three), medullary thyroid carcinoma (two), thymic carcinoid (two), and a bronchial carcinoid tumor (one). The re- maining ten patients (48%) are alive with a mean fol- low-up of 50 months; nine are disease free with one pa- tient (with bronchial carcinoid) manifesting metastatic disease 74 months after total adrenalectomy.
SURGICAL RESULTS AND APPROACH
Seventy-three (80%) of the ninety-one patients were explored via the posterior approach, and eighteen patients were explored anteriorly. Fifty of the ninety- one patients underwent bilateral total adrenalectomy. Two patients (2.2%) died within 30 days of operation. One patient, who required open drainage of a lesser sac abscess, had a wound infection. No instances of deep venous thrombosis or pulmonary embolism occurred during their hospitalization. Both deaths occurred after hospital dismissal. The first occurred 3 weeks after op-
eration in a 71-year-old woman with ACTH-indepen- dent adrenocortical hyperplasia. Although the cause of death was indeterminate, some question remains re- garding her compliance with replacement therapy. The second death, in a 67-year-old man with ectopic ACTH production, was from an extensively metastatic adeno- carcinoma of the prostate. At the time of his urgent adrenalectomy he had systemic candidiasis; after oper- ation he experienced multisystem organ failure, which caused his death. Serious postoperative morbidity was confined to one patient in whom developed a left sub- phrenic abscess that required surgical drainage after a bilateral anterior adrenalectomy. Delayed wound heal- ing occurred in three patients, one of whom had the in- fected wound mentioned above.
The mean length of hospitalization for patients un- dergoing unilateral anterior, bilateral anterior, unilat- eral posterior, and bilateral posterior adrenalectomy was 8.0, 11.0, 6.0, and 6.0 days, respectively.
The perioperative use of vitamin A, prophylactic an- tibiotics, and anticoagulation was not uniform. These adjunct measures were used in 8.8%, 60.4%, and 5.5% of the patients, respectively. To evaluate the incidence of delayed wound infection and delayed wound healing, the 71 patients who were alive at the completion of the study were all contacted personally. In these 71 patients three instances of delayed wound healing, which re- quired observation only, were encountered. One patient had small amounts of serosanguineous fluid draining from the incision for approximately 1 month after adre- nalectomy.
PATHOLOGIC ASPECTS
All adrenal glands were weighed immediately after excision. Mean combined adrenal weight in patients with Cushing’s disease was 15.5 grams in comparison with 20.1 grams in those with ectopic ACTH-produc- ing tumors. The mean combined weight in the group of patients with bilateral macronodular hyperplasia was 119.85 grams, as opposed to 18.25 grams in those with micronodular disease.
In the group of patients with Cushing’s syndrome, the
mean diameter of removed hyperfunctional benign adrenal adenomas (34 patients) was 3.9 centimeters in contrast with a mean diameter of 7.6 centimeters for the four patients with primary adrenocortical carcino- mas.
DISCUSSION
More than 40 years ago Priestley et al.5 succinctly documented the initial Mayo Clinic experience with subtotal adrenalectomy for Cushing’s syndrome. Care- ful review of this landmark presentation before the American Surgical Association gives us an improved perspective of our current approach to patients with these problems. Priestley reported on twenty-nine pa- tients, all of whom appeared to have had Cushing’s dis- ease with bilateral adrenocortical hyperplasia and all of whom were treated by total unilateral and contralateral subtotal (90%) adrenal resection, usually as a staged surgical procedure. From those early days of the eluci- dation of replacement steroid therapy it is fascinating to read: “After resection of the second adrenal gland, the patients usually did well for ten to 20 days, and then a delayed reaction, characterized by anorexia, increasing nausea and finally vomiting developed. The anorexia frequently progressed until the patients had such an ac- tual aversion to food that the mere sight of food would result in vomiting. Weakness, fever and tachycardia were usually present but the blood pressure as a rule did not fall to the low levels usually observed in acute ad- renal insufficiency.” “The chemical composition of the blood gradually became disturbed and the 24-hour ex- cretion of 17-ketosteroids and corticosteroids reached low levels.” “In the last nine cases of the series, corti- sone has been used in place of aqueous adrenal cortical extract … Since the institution of this program, the immediate postoperative shocklike state has been avert- ed.”
In this initial report six (20.7%) of the 29 patients died during the early postoperative period; some 40 years later, as reported in the current study, it is pleas- ing to note that this figure has dropped to zero.
Twenty years after the report of Priestley et al.,5 Bennett et al.6 detailed a similar experience from the Peter Brent Brigham Hospital in Boston. In this report the immediate postoperative mortality had decreased to 5.2%, whereas fully 23.1% of patients experienced postoperative wound infection. More recent reviews from Sweden,7 Norway,8 and Denmark9 reported post- operative mortality rates of 0%, 9.1%, and 8.0%, respectively, with concomitant wound infection rates of 4%, 8%, and 10.8%, respectively. In the United States Sarkar et al.10 in 1990 evaluated 44 patients undergo- ing adrenalectomy for Cushing’s syndrome. In this se- ries the operative mortality rate was 2.3% and the
wound infection rate was 4.5%. In a smaller series Bru- nicardi et al.11 reported an operative mortality rate of 14% with impaired wound healing in 43% of their pa- tients. They postulated that because wound dehiscence typically occurred on the concave lateral aspect of their Hugh-Young incisions, poor regional tissue perfusion may have played a role to account for this high incidence of wound healing problems.
In the classic text Boyd’s Pathology for the Surgeon12 the deleterious effect of steroids on wound healing are noted: “Cortisone and ACTH delay the process of re- pair in the experimental animal. There is a decrease in the intensity of the vascular and exudative changes and an almost complete lack of vascular and fibroblastic proliferation and migration of histiocytes.” A number of investigators13-15 have clearly demonstrated that the antiinflammatory action of steroids leads to a decrease in the breaking strength of wounds, interference with wound contraction, and delayed reepithelialization and angiogenesis. Although there are repeated implications in the surgical and medical literature that hypercortiso- lism (endogenous or exogenous) will result in a high in- cidence of surgical wound infections and delayed wound healing, this fact is not substantiated in most recent re- ports including the current study. An interesting obser- vation is the effect of vitamin A on wound healing. The administration of vitamin A has no effect on wound healing in the patient with normal steroid levels; how- ever, in the steroid-inhibited wound vitamin A (25,000 IU/day) has been shown to accelerate wound strength gain and reepithelialization.16, 17 We have seldom used (8.8%) vitamin A in our patients with hypercortisolism; the negligible incidence of wound problems appears to justify this current approach.
There have been suggestions in the literature18, 19 that the seeming increase in postoperative venous thrombotic episodes in patients with hypercortisolism and the com- plications thereof were due to an increase in the level of factor VIII leading to a hypercoagulable state. Dal Bo Zanon et al.18 demonstrated a uniform elevation in fac- tor VIII levels in 15 patients with Cushing’s syndrome, levels that normalized 3 to 4 months after operation. Sjöberg et al.19 found postoperative thromboembolic oc- currences in seven of 19 patients with Cushing’s syndrome and demonstrated a positive correlation between the increase in factor VIII and the severity of the hypercortisolism. Both of these authorities raised the issue of prophylactic heparinization in patients undergoing operation for hypercortisolism. We have used this approach sparingly (5.5% in this study) and have relied instead on fairly routine use of intraopera- tive sequential venous compression of the patients’ legs, expeditious adrenalectomy, and early and vigorous am- bulation.
The treatment of choice for Cushing’s disease contin- ues to be transsphenoidal adenomectomy. The success rate for this neurosurgical procedure ranges from 75% to 96%.20-22 During the period encompassed by our current study selective transsphenoidal adenomectomy was curative in 202 patients with pituitary-dependent Cushing’s syndrome. However, 24 patients (10%) were not cured with this operation because of incomplete re- section of a microadenoma (46%), an inability to locate a small microadenoma (33%), or incomplete resection of an invasive macroadenoma (21%). In one third of these patients developed Nelson-Salassa syndrome with hy- perpigmentation and pituitary macroadenoma. This consequence was avoided in some patients with pitu- itary radiation therapy (13 of 24 patients).
The cause of adrenal-dependent Cushing’s syndrome in this study was not unusual, with the majority (76%) of patients harboring benign adrenocortical adenomas. An intriguing aspect is that the exact cause of ACTH- independent adrenocortical hyperplasia remains un- known and may be of either the macronodular or micronodular (PPNAD) type. Patients with macro- nodular hyperplasia typically have large adrenal glands as evident by the average combined adrenal weight of 120 grams in this study. The finding of PPNAD should alert the clinician to the possible presence of the Carney complex (PPNAD, cardiac myxomas, pigmented skin lesions, and peripheral nerve tumors)23; this complex was found in one patient in this study. Our findings in this subgroup of patients with hypercortisolism is sim- ilar to those recently reported from the National Insti- tutes of Health by Perry et al.24
An equally fascinating and challenging subgroup of patients are those with ectopic (nonpituitary) sources of ACTH production. The treatment of choice in these patients is resection of the primary tumor, with bilateral adrenalectomy reserved only for those patients with un- resectable primary tumors, metastatic disease, or those in whom the primary site is unknown. In our 21 patients only one patient underwent successful resection of an ectopic source of both ACTH and corticotropin-releas- ing factor production (pheochromocytoma). Metastatic- recurrent or unresectable disease was present in 11 pa- tients with the remaining nine patients having unknown ectopic sources. Five (24%) of the patients with ectopic ACTH syndrome underwent prior transsphenoidal ex- ploration for suspected pituitary-dependent disease be- fore discovery of the ectopic nature of their disorder and the curative bilateral adrenalectomy. In some patients with ectopic ACTH syndrome the cause of their disease is obvious, because it is characterized by rapid onset, progressive clinical course, and a radiographically evi- dent neoplasm. However, in most patients with ectopic
ACTH syndrome (e.g., bronchial carcinoid) the clinical course and results on dynamic endocrine testing may be indistinguishable from patients with pituitary-depen- dent disease.3 In 93% of the patients with ACTH- secreting bronchial carcinoid tumors reported in the literature, the initial clinical manifestations were con- sistent with typical pituitary-dependent disease and in 60% the neoplasm was initially occult.3 Inferior petro- sal sinus sampling for ACTH with oCRH stimulation may be the only test to reliably distinguish pituitary- dependent disease from an ectopic source of ACTH. This test has been used at our institution since 1986 and has reduced inappropriate transsphenoidal explorations since that time.
Surgical cure of Cushing’s syndrome results in acute adrenocortical insufficiency. To prevent the postopera- tive shocklike state described by Priestley et al.5 in 1951, we routinely use high-dose glucocorticoid coverage. Our standard glucocorticoid preparation for surgery is 40 mg methylprednisolone sodium succinate (Solu Medrol; Upjohn, Kalamazoo, Mich.) administered intramuscu- larly in the morning of the operation and again in the evening of the operation; the dose is tapered to 20 mg and 10 mg intramuscularly every 12 hours on postoperative days 2 and 3, respectively. The patient’s clinical condi- tion guides how much further to taper the dose to par- allel clinical improvement and when to initiate orally administered glucocorticoid. Most patients have been exposed to endogenous hypercortisolism for many years, and abrupt postoperative institution of replacement doses of glucocorticoids results in significant withdrawal symptoms. Typically, patients are dismissed from the hospital on prednisone (Deltasone; Upjohn) 5 mg orally (or the equivalent dose of hydrocortisone) in the morn- ing and 5 mg orally before the evening meal; 1 to 2 weeks later the dosage is decreased to 5 mg in the morning and 2.5 mg in the afternoon.
In patients treated with bilateral adrenalectomy, prednisone or an equivalent of hydrocortisone at a dosage of 5 mg to 7.5 mg daily is continued lifelong. In many cases the full replacement dosage is not needed.25 We instruct patients on stress steroid coverage and the need to wear medical identification. In addition, orally administered mineralocorticoid replacement in the form of fludrocortisone (Florinef; Apothecon, Princeton, N.J.) at a dosage of 0.05 to 0.1 mg per day is started before dismissal from the hospital. The proper dosage (0.05 to 0.2 mg daily) is determined by ser- um electrolytes level and supine and standing blood pressures.
A recent report from Northern Ireland26 corroborates the results of our current study and emphasizes the safety of bilateral adrenalectomy. In this report of 26
patients with Cushing’s disease undergoing bilateral adrenalectomy, the operative mortality rate was zero, one patient had a wound infection, and two additional patients experienced either a pulmonary embolus or a subphrenic abscess, respectively.
In patients treated with unilateral adrenalectomy it may take 4 to 12 months before the hypothalamic-pitu- itary-adrenal axis recovers.27 More recent studies show that the rate-limiting step in axis recovery is in the hy- pothalamic corticotropin-releasing hormone neuron and/or higher regulatory inputs.28 At approximately 6 weeks after operation patients are switched to a single morning dose of glucocorticoid. The single morning dose allows for relatively nocturnal glucocorticoid insuffi- ciency and stimulation of the hypothalamic neurons. The concentration of plasma cortisol at 8 AM (before glucocorticoid administration) is measured every 6 to 8 weeks. Glucocorticoid treatment may be discontinued when the plasma cortisol concentration is 10 ug/dl or greater. Patients are cautioned on the need for stress steroid coverage for approximately 1 year after exoge- nous treatment is discontinued. Mineralocorticoid re- placement is usually not required in patients after uni- lateral adrenalectomy.
An interesting finding in this study was the predom- inance of women in all three categories: Cushing’s dis- ease, 76%; adrenal-dependent Cushing’s syndrome, 91%; and ectopic ACTH-producing hypercortisolism, 62%. Although female predominance is typical of pitu- itary-dependent Cushing’s syndrome, this gender pre- dominance for the other subtypes has not been empha- sized before.
We continue to favor the posterior approach to the adrenal glands (with the exception of patients with adrenocortical cancer or with tumors greater than 5 cm in diameter), an approach used in 80% of the patients in this study. Although technically demanding at times, the safety of this approach with its minimal morbidity and mortality fully justifies its continued widespread utilization. Thus far our clinical experience with lap- aroscopic adrenalectomy is limited. The laparoscopic removal of the adrenal glands is in its infancy and should be carefully evaluated by all endocrine surgeons. The standard against which this possible new mode of adre- nalectomy for hypercortisolism should be measured is, we believe, the data contained in this study.
In 1855 Thomas Addison stated: “It will hardly be disputed that at the present moment, the function of the supra-renal capsules, and the influence they exercise in the general economy are almost or altogether un- known.” Although we yet have much to learn, we have come a long way in the management of those often crit- ically ill and fragile patients.
REFERENCES
1. Kaye TB, Crapo L. The Cushing’s syndrome: an update on di- agnostic tests. Ann Intern Med 1990;112:434-44.
2. Flack MR, Oldfield EH, Cutler BG Jr, et al. Urine free cor- tisol in the high-dose dexamethasone suppression test for the differential diagnosis of the Cushing syndrome. Ann Intern Med 1992;116:211-7.
3. Leinung MC, Young WF Jr, Whitaker MD, Scheithauer BW, Trastek VF, Kvols LK. Diagnosis of corticotropin-producing bronchial carcinoid tumors causing Cushing’s syndrome. Mayo Clin Proc 1990;65:1314-21.
4. Oldfield EH, Doppman JL, Nieman LK, et al. Petrosal sinus sampling with and without corticotropin-releasing hormone for the differential diagnosis of Cushing’s syndrome. N Engl J Med 1991;325:897-905.
5. Priestley JT, Sprague RG, Walters W, Salassa RM. Subtotal adrenalectomy for Cushing’s syndrome: a preliminary report of 29 cases. Ann Surg 1951;134:464-72.
6. Bennett AH, Cain JP, Dluhy RG, Tynes WV, Harrison JH, Thorn GW. Surgical treatment of adrenocortical hyperplasia: 20-year experience. J Urol 1973;109:321-4.
7. Malmaeus J, Markaes A, Oberg K, et al. Adrenal gland surgery: preoperative location of lesions, histologic findings and outcome of surgery. Acta Chir Scand 1986;152:577-81.
8. Grabner P, Hauer-Jensen M, Jervell J, Flatmark A. Long-term results of treatment of Cushing’s disease by adrenalectomy. Eur J Surg 1991;157:461-4.
9. Blichert-Toft M, Bagerskov A, Lockwood K, Hasner E. Oper- ative treatment, surgical approach, and related complications in 195 operations upon the adrenal gland. Surg Gynecol Obstet 1972;135:261-6.
10. Sarkar R, Thompson NW, McLeod MK. The role of adrena- lectomy in Cushing’s syndrome. SURGERY 1990;108:1079-84.
11. Brunicardi FC, Rosman PM, Lesser KL, Andersen DK. Cur- rent status of adrenalectomy for Cushing’s disease. Surgery 1985;98:1127-1133.
12. Anderson W, ed. Boyd’s pathology for the surgeon. 8th ed. Philadelphia: WB Saunders, 1967.
13. Hinshow DB, Hughes LD, Stafford CE. Effects of cortisone on the healing of disrupted abdominal wounds. Am J Surg 1961; 101:189-91.
14. Salmela KI, Ahonen J. The effect of methylprednisolone and vitamin A on wound healing. Acta Chir Scand 1981;147:307-12.
15. Stephens FD, Dunphy JE, Hunt TK. The effect of delayed ad- ministration of corticosteroids on wound contraction. Ann Surg 1971;173:214-8.
16. Ehrlich HP, Hunt TK. Effects of cortisone and vitamin A on wound healing. Ann Surg 1968;167:324-8.
17. Hunt TK, Ehrlich HP, Garcia JA, Dunphy JE. Effect of vi- tamin A on reversing the inhibitory effect of cortisone on heal- ing of open wounds in animals and man. Ann Surg 1969;170:633- 41.
18. Dal Bo Zanon R, Fornasiero L, Boscaro M, Cappellato G, Fabris F, Birolami A. Increased factor VIII associated activities in Cushing’s syndrome: a probable hypercoagulable state. Thromb Haemost 1982;47:116-7.
19. Sjöberg HE, Blombäck M, Granberg PO. Thromboembolic complications, heparin treatment and increase in coagulation factors in Cushing’s syndrome. Acta Med Scand 1976;199:95-8.
20. Chandler WF, Schteingart DE, Lloyd RV, et al. Surgical treat- ment of Cushing’s disease. J Neurosurg 1987;66:204-12.
21. Fahlbusch R, Buchfelder M, Muller OA. Transsphenoidal sur- gery for Cushing’s disease. J R Soc Med 1986;79:262-9.
22. Salassa RM, Laws ER Jr, Carpenter PC, et al. Transsphenoi- dal removal of pituitary microadenoma in Cushing’s disease. Mayo Clin Proc 1978;53:24-8.
23. Carney J, Gordon H, Carpenter P, et al. The complex of myx- omas, spotty pigmentation and endocrine overactivity. Medicine 1985;64:270-83.
24. Perry RR, Nieman LK, Cutler GB Jr, et al. Primary adrenal causes of Cushing’s syndrome: diagnosis and surgical manage- ment. Ann Surg 1989;210:59-68.
25. Kemink L, Hermus A, Pieters G, Benraad TH, Smals A, Klop- penborg P. Residual adrenocortical function after bilateral adrenalectomy for pituitary-dependent Cushing’s syndrome. J Clin Endocrinol Metab 1992;75:1211-4.
26. McCance DR, Russell CFJ, Kennedy TL, Hadden DR, Kennedy L, Atkinson AB. Bilateral adrenalectomy: low mor- tality and morbidity in Cushing’s disease. Clin Endocrinol 1993; 39:315-21.
27. Graber AL, Ney RL, Nicholson WE, Island DP, Liddle GW. Natural history of pituitary-adrenal recovery following long- term suppression with corticosteroids. J Clin Endocrinol Metab 1965;25:11-6.
28. Gomez MT, Magiakou MA, Mastorakos G, Chrousos GP. The pituitary corticotroph is not the rate limiting step in the postop- erative recovery of the hypothalamic-pituitary-adrenal axis in patients with Cushing Syndrome. J Clin Endocrinol Metab 1993;77:173-7.
BOUND VOLUMES AVAILABLE TO SUBSCRIBERS
Bound volumes of SURGERY are available to subscribers (only) for the 1995 issues from the publisher at a cost of 100.58 for Canadian, and $94.00 for international sub- scribers for Vol. 117 (January to June) and Vol. 118 (July to December). Shipping charges are in- cluded. Each bound volume contains a subject index and an author index, and all advertising is re- moved. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular journal subscription. Payment must accompany all orders. Contact Mosby-Year Book, Inc., Subscription Services, 11830 Westline Industrial Dr., St. Louis, MO 63146-3318; phone (800)325- 4177, ext. 4351, or (314)453-4351.