PRIMARY ALDOSTERONISM AND MALIGNANT ADRENOCORTICAL NEOPLASIA
ROBERT M. SALASSA, M.D. AND (by invitation) RICHARD E. WEEKS, M.D., ROBERT C. NORTHCUTT, M.D. AND J. AIDAN CARNEY, M.D.
ROCHESTER, MINNESOTA
Primary aldosteronism is usually defined as a disorder of the adrenal cortex resulting from excessive and inappropriate secretion of the mineralo- corticoid, aldosterone, but without either clinical or chemical evidence of other abnormalities of adrenocortical function. The vast majority of pa- tients with primary aldosteronism show little on clinical examination other than hypertension. Thus the diagnosis depends largely on appropriate laboratory investigation of the hypertensive patient and the finding of (1) nonsuppressible elevated secretion or excretion of aldosterone, (2) sup- pressed plasma renin activity in spite of acute volume depletion and up- right posture, and (3) hypokalemia when on a generous sodium intake.
The initial case reported by Conn1, 2 20 years ago and most of the early reported cases3, 4 were found to have small benign adrenal tumors. How- ever, it soon became evident that not all cases were due to a single small adenoma of the adrenal cortex. The same clinical and laboratory abnor- malities were noted in patients who had multiple small nodules-so-called nodular or adenomatous hyperplasia5-9-and in many series this consti- tutes 20 to 30% of patients with primary aldosteronism. This brief pres- entation concerns a third uncommon cause of primary aldosteronism, malignant adrenocortical neoplasia.
In 1955 Foye and Feichtmeir10 reported on a patient with an adrenocor- tical carcinoma who presented with the clinical features of primary aldos- teronism. Since this paper there have been occasional reports of single cases of adrenocortical carcinoma presenting with the clinical features of mineralocorticoid excess but without clinical features of glucocorticoid or androgen excess. About 15 cases have been reported or at least referred to in the literature.3, 10-23 In the case reported by Foye and Feichtmeir,10 and in many of the cases subsequently reported, there was chemical evi-
From the Division of Endocrinology and Internal Medicine, Mayo Clinic and Mayo Foundation, Rochester, Minnesota; From the Department of Surgical Pathology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota.
Reprint address: Dr. R. M. Salassa, Mayo Clinic, 200 First Street SW, Rochester, MN 55901.
Presented at the meeting of the American Clinical and Climatological Association, Williamsburg, VA, October 22 to 24, 1974.
dence of additional adrenocortical dysfunction such as excessive secretion of glucocorticoids or androgens (or both) despite the absence of clinical evidence of excess glucocorticoid or androgen activity. In other cases the excretion of other steroids such as desoxycorticosterone, corticosterone, and their urinary metabolites was elevated or aldosterone was not meas- ured.
Thus, most of these reported cases of adrenocortical carcinoma present- ing with only the clinical features of mineralocorticoid excess do not, in the strict sense, meet the definition of primary aldosteronism. Some authors have stated that patients with adrenal carcinoma presenting as primary aldosteronism can be identified with confidence by the presence of steroid abnormalities other than excessive aldosterone secretion. In general these patients have had large tumors, varying in weight from 30 g up to 2,020 g. As a group, these patients have had a poor prognosis-more than half were dead at the time that their cases were reported.
CASES STUDIED
During the past 17 years we have operated on 105 patients presenting with primary aldosteronism; 6 were found to have large adrenocortical tumors, weighing from 45 to 1,050 g with an average of 472 g. The initial pathologic report in each of these cases was “adrenocortical carcinoma.” A later and more critical review indicated that the four largest tumors (250, 260, 900, and 1,050 g) had sufficient classical features of malignant neoplasia, as illustrated in Table 1, that the histologie diagnosis of carci- noma was unequivocal. The diagnosis of carcinoma was cquivocal in the two small tumors (45 and 60 g). Since the pathologist cannot always dis- tinguish between large benign adenomas and low-grade adenocarcinomas arising in the adrenal cortex, it seemed reasonable to compare the clinical and laboratory data on these six cases with similar data from the other cases in our series. Sixty-seven of these other cases were due to single ade- nomas and 32 cases had multiple adenomas or nodules and were classified as adenomatous hyperplasia.
There was nothing unique or distinctive about the sex distribution, the age, or the hypertension of the 6 patients with the large tumors compared to the other 99 patients in the series. One of the six patients had group 4 hypertension, and one of the patients with adenomatous hyperplasia also had accelerated hypertension (group 3). The complaints of polyuria and muscular weakness were more prominent among the six patients with the malignant and possibly malignant tumors, and two of these six patients had had generalized seizures, probably reflecting a more severe degree of electrolyte disturbance in this group compared to the rest of the series.
As illustrated in Table 2, the 6 cases, as a group, tended to have higher
| Carcinoma (4 cases) | "Adenoma" (2 cases) | |
|---|---|---|
| Gross | ||
| Encapsulated. | Yes, thick | Yes, fine |
| Necrosis | Yes (4) | No |
| Hemorrhage | Yes (4) | No |
| Lobulation | Yes, coarse | No |
| Microscopic | ||
| Capsular invasion | Yes (4) | No |
| Capsular penetration | Yes (1) | No |
| Vascular invasion | Yes (3) | No |
| Solid pattern | Yes (4) | No |
| Trabecular pattern | No | Yes (2) |
| Mitosis. | Frequent | Rare |
| Bizarre giant cells | Yes (2) | No |
| Reticular cell type | Yes (4) | No |
| Fascicular cell type | No | Yes (2) |
| Adrenocortical disease | No. of pt | % of patients with *: | |||
|---|---|---|---|---|---|
| K > 2.5 A < 30 | K > 2.5 A ≥ 30 | K ≤ 2.5 A < 30 | K ≤ 2.5 A ≥ 30 | ||
| Hyperplasia | 32 | 81 | 13 | 6 | 0 |
| Single adenoma | 67 | 42 | 24 | 12 | 22 |
| Malignant | 6 | 17 | 17 | 0 | 66 |
* K (potassium), meq/liter; A (aldosterone), ug/24 h.
levels of urinary aldosterone and lower levels of serum potassium when com- pared to the 67 cases due to solitary adenomas, and especially when com- pared to the 32 cases with adenomatous hyperplasia. There was no signifi- cant difference in the levels of plasma renin activity among the three groups. Pertinent steroid data on the six cases are presented in Table 3. The uri- nary excretion of aldosterone was elevated in each patient and the urinary excretion of 17-ketosteroids and 17-ketogenic steroids and the plasma cortisol levels were normal in five of the cases. Tetrahydro-11-deoxycortisol was measured in three patients and was elevated in one. In this one pa- tient with the elevated urinary tetrahydro-11-deoxycortisol there was also an elevation of urinary 17-ketogenic steroids, undoubtedly because of the
| Case | Urinary | Plasma 17-OHCS§ (mg/dl) 8 a.m./4 p.m. | |||
|---|---|---|---|---|---|
| Aldosterone (µg/24 h) | 17-KS* (mg/24 h) | 17-KGSt (mg/24 h) | THSt (mg/24 h) | ||
| 1| | 46 | 3.9 | 7.6 | 15.3/12.6 | |
| 2 | 58 | 12.2 | 24.0 | 8.6 | 17.2/12.6 |
| 3 | 58 | 7.9 | 11.8 | 0.1 | 14.0/5.8 |
| 4| | 26 | 3.9 | 5.3 | 16.2/7.1 | |
| 5 | 211 | 8.4 | 5.4 | 0.2 | 14.1/6.6 |
| 6 | 28 | 6.4 | 4.4 | 23.4/9.3 | |
* 17-Ketosteroids.
+ 17-Ketogenic steroids.
# Tetrahydro-11-deoxycortisol.
§ 17-Hydroxycorticosteroids.
|| Questionable malignancy.
increased urinary excretion of tetrahydro-11-deoxycortisol. The plasma cortisol levels were normal in this patient, as was true in the other five cases. Figure 1 illustrates the large (1,050 g) carcinoma in patient 3. Table 4 shows preoperative and postoperative steroid data before and after at- tempted suppression with dexamethasone and fludrocortisone in this pa- tient. There was no evidence of adrenocortical dysfunction other than the inappropriate and nonsuppressible excessive aldosterone excretion. In general, there was no chemical evidence of excessive glucocorticoid or an- drogen production in any of the six patients. However, one patient did have chemical evidence of additional adrenocortical dysfunction as indi- cated by elevated urinary levels of tetrahydro-11-deoxycortisol.
Attempts were made to visualize the tumors in these six patients. Five of the tumors were visualized by either excretory urography or bolus nephrotomography on the initial attempt. However, in patient 5 both the excretory urogram and the nephrotomograms were initially negative but were positive 16 months later. Two of the six patients received iodocho- lesterol in an effort to visualize their tumors by external scanning. How- ever, in neither case was the tumor definitely visualized.
As was the case with patients reported in the literature, our patients with unequivocally malignant tumors have had a poor prognosis. Two have died of their disease, one is living with metastases, and only one is alive and well without clinical or biochemical evidence of recurrence 11 years after sur- gery. The two patients with large tumors that had an equivocal pathologic diagnosis of adrenocortical carcinoma are alive and well without clinical or chemical evidence of recurrence 17 and 9 years following surgery.
MIKE 2 3 4 5 6 7 8 9 10 11 12 18 14 815
The course of one patient (case 5) was particularly instructive. He was a 52-year-old farmer with insulin-dependent diabetes mellitus of 25 years’ duration, moderately severe diabetic retinopathy, grade 2 albuminuria, grade 2 microhematuria, and hypertension. Before his admission the pa- tient had been seriously ill for 2 weeks, requiring hospitalization elsewhere and the use of several medications and intravenous fluids for his hyperten- sion, congestive heart failure, and electrolyte abnormalities, including se- vere hypokalemia.
Several days after admission to our institution, the urinary aldosterone excretion was high (250 ug/24 h); a morning determination of peripheral plasma renin activity, obtained in the supine position without sodium de- pletion, was 19.8 ng/liter per min. The excretory urogram and nephrotomo- gram were negative. About 3 weeks later, when the patient’s clinical con- dition had improved and he had lost 20 pounds of fluid, the urinary excre- tion of aldosterone was still high (211 ug/day), but a determination of peripheral plasma renin activity when the patient was sodium-depleted and upright was only 2.6 ng/liter per min. Because of the negative ex- cretory urogram and nephrotomogram, the diagnosis of a large, possibly
| Urinary | Plasma 17-OHCS§ (µg/dl) 8 a.m./4 p.m. | Treatment|| | ||||
|---|---|---|---|---|---|---|
| 17-KS* (mg/24 h) | 17-KGSt (mg/24 h) | Aldo- sterone (µg/24 h) | THS# (mg/24 h) | |||
| Preop. | 7.9 | 11.8 | 44 | 0.2 | 14.0/5.8 | |
| 6.9 | 3.0 | 58 | 1.0/2.5 | Dex. (2 mg) | ||
| 40 | FF (1 mg) | |||||
| Postop. | ||||||
| 6 days | 0.5 | |||||
| 18 mo | 7.7 | 11.0 | 7.0 | |||
| 21/2 yr | 8.6 | 8.2 | 5.0 | 12.9/6.0 | ||
| 51/2 yr | 4.3 | 7.2 | 20.0 | 0 | 14.9/11.9 | |
| 2.0 | 0.2 | 12.0 | 1.6/3.7 | Dex. (2 mg) | ||
| 4.1 | 5.7 | 2.0 | FF (1 mg) | |||
| 71/2 yr | 8.2 | 6.2 | 14.0 | |||
| 7.4 | 8.7 | 4.0 | FF (1 mg) | |||
| 10 yr | 9.7 | 5.6 | 13.0 | 0 | ||
* 17-Ketosteroids.
+ 17-Ketogenic steroids.
# Tetrahydro-11-deoxycortisol.
§ 17-Hydroxycorticosteroids.
| Dex. = dexamethasone; FF = fludrocortisone.
malignant, adrenal tumor seemed unlikely, and he was started on spirono- lactone, 400 mg/day.
He improved rapidly and returned to work. For about 1 year he continued to do well and worked without difficulty. Then muscular weakness recurred. When he returned for reevaluation 16 months after the initial visit, he had a positive excretory urogram and pulmonary and liver metastases from a left adrenal carcinoma (found to weigh 900 g).
SUMMARY
Our experience indicates that although adrenal carcinoma is not a com- mon cause of primary aldosteronism, 4 to 5% of patients in a single large series may have a malignant adrenocortical tumor. The magnitude of the hypokalemia and the hyperaldosteronuria tends to be greater in patients with malignant tumors, but these patients cannot be clearly separated from those with benign tumors or hyperplasia on this basis. Patients who have malignant tumors may have no chemical evidence of adrenocortical dysfunction other than excessive aldosterone secretion. Finally, a good re-
sponse to spironolactone for months does not exclude adrenal carcinoma as the cause of primary aldosteronism.
REFERENCES
1. CONN, J. W .: Presidential Address. Part II. Primary Aldosteronism, a New Clini- cal Syndrome. J. Lab. Clin. Med. 45: 6, 1955.
2. CONN, J. W .: Primary Aldosteronism. J. Lab. Clin. Med. 45: 661, 1955.
3. CONN, J. W., KNOPF, R. F., AND NESBIT, R. M .: Clinical Characteristics of Pri- mary Aldosteronism From an Analysis of 145 Cases. Am. J. Surg. 107: 159, 1964.
4. CONN, J. W .: The Evaluation of Primary Aldosteronism 1954-1967. Harvey Lect. 62: 257, 1966-1967.
5. LARAGH, J. H., LEDINGHAM, J. G. G., AND SOMMERS, S. C .: Secondary Aldosteron- ism and Reduced Plasma Renin in Hypertensive Disease. Trans. Assoc. Am. Physicians 80: 168, 1967.
6. DAVIS, W. W., NEWSOME, H. H., WRIGHT, L. D., HAMMOND, W. R., EASTON, J., AND BARTTER, F. C .: Bilateral Adrenal Hyperplasia as a Cause of Primary Al- dosteronism With Hypertension, Hypokalemia and Suppressed Renin Activity. Am. J. Med. 42: 642, 1967.
7. KATZ, F. H .: Primary Aldosteronism With Suppressed Plasma Renin Activity Due to Bilateral Nodular Adrenocortical Hyperplasia. Ann. Intern. Med. 67: 1035, 1967.
8. LIDDLE, G. W .: Discussion. Trans. Assoc. Am. Physicians 80: 179, 1967.
9. BIGLIERI, E. G., SCHAMBELAN, M., SLATON, P. E., AND STOCKIGT, J. R .: The Inter- current Hypertension of Primary Aldosteronism. Circ. Res. 27 Suppl. 1: 195, 1970.
10. FOYE, L. V., AND FEICHTMEIR, T. V .: Adrenal Cortical Carcinoma Producing Solely Mineralocorticoid Effect. Am. J. Med. 19: 966, 1955.
11. BROOKS, R. V., McSWINEY, R. R., PRUNTY, F. T., AND WOOD, F. J .: Potassium Deficiency of Renal and Adrenal Origin. Am. J. Med. 23: 391, 1957.
12. BLEHA, O., POLAK, E., AND MOTLIK, K .: Primarne aldosteronismus II [Primary Aldosteronism II]. Sb. Lek. 9: 259, 1957.
13. KANDRAČ, M. Š., ZIKANOVA, Z. D., AND MACH, O .: Aldostern a kortikosteroidy u nemocné primárním aldosteronismem a s karcinomem kůry nadledvinek [Aldos- terone and Corticosteroids in a Patient Suffering From Primary Aldosteronism and Carcinoma of the Adrenal Cortex]. Sb. Lek. 9: 283, 1957.
14. ZIMMERMAN, B., MORAN, W. H., JR., ROSENBERG, J. C., KENNEDY, B. J., AND FREY, R. J .: Physiologic and Surgical Problems in the Management of Primary Aldosteronism. Ann. Surg. 150: 653, 1959.
15. FELDMAN, S., AND RAVERA, J. J .: Hipertensión arterial por tumor maligno primi- tivo córticoadrenal. Torax. 10: 284, 1961.
16. CRANE, M. G., HARRIS, J. J., AND HERBER, R .: Primary Aldosteronism Due to an Adrenal Carcinoma. Ann. Intern. Med. 63: 494, 1965.
17. SANTANDER, S. R., GONZÁLEZ, A., AND SUÁREZ, J. A .: Case of Probable Mineralo- corticoid Excess Without Hypercortisolism Due to a Carcinoma of the Adrenal Cortex. J. Clin. Endocrinol. Metab. 25: 1429, 1965.
18. MARQUEZY, R. A., BRICAIRE, H., LAUDAT, M. H., COURJACKET, J., AND PHILBERT, M .: Adenocarcinoma of the Adrenal With Syndrome of Hyperandrogenia and Hypermineralocorticism. Ann. Endocrinol. (Paris) 26: 247, 1966.
19. NEVILLE, A. M., AND SYMINGTON, T .: Pathology of Primary Aldosteronism. Can- cer 19: 1854, 1966.
20. HARRISON, J. H .: Discussion. Ann. Surg. 164: 609, 1966.
21. ALTERMAN, S. L., DOMINGUEZ, C., LOPEZ-GOMEZ, A., AND LIEBER, A. L. : Primary Adrenocortical Carcinoma Causing Aldosteronism. Cancer 24: 602, 1969.
22. BROOKS, R. V., FELIX-DAVIES, D., LEE, M. R., AND ROBERTSON, P. W .: Hyperal- dosteronism From Adrenal Carcinoma. Br. Med. J. 1: 220, 1972.
23. FILIPECKI, S., FELTYNOWSKI, T., POPŁAWSKA, W., ŁAPÍNSKA, K., KRUŚ, S., WOCIAL, B., AND JANUSZEWICZ, W .: Carcinoma of the Adrenal Cortex With Hy- peraldosteronism. J. Clin. Endocrinol. Metab. 35: 225, 1972.
DISCUSSION
DR. GRANT W. LIDDLE (Nashville): I should like to congratulate Dr. Salassa for his very clear presentation of a beautiful study. Dr. Salassa, I wonder if you have had an opportunity to treat any of the metastatic carcinomas with Op’DDD or Mito- tane? This is the only drug I know of which has any influence on the growth of adrenal cortical carcinomas. There is perhaps a special reason for being pessimistic about whether it would be effective in aldosterone producing tumors since in normal ad- renals the aldosterone-producing cells seem to be less sensitive to destruction by Op’DDD than do the cortisol-secreting cells. I wonder if you have had any oppor- tunity at all to study this drug?
DR. SALASSA: In answer to your first point, we have had one man who really did quite well for about 14 months on Op’DDD after he developed inoperable metastases. By quite well, I mean his hypertension was ameliorated but not really returned to normal. It was down within an acceptable range with diastolic levels varying between 90 and 100 and he was also normokalemic and the elevated urinary aldosterone fell to within the normal range. We wondered if the drug would be effective when we started the patient on Op’DDD). However, if you noticed the first slide, in our particular ex- perience, the cells of these malignant tumors resembled more the reticularis type of cell and it is possible that these cells and tumors rise from the reticularis portion of the adrenal cortex and maybe this is the explanation for the response. At any rate, one patient we tried it on did relatively well. There is another patient who is on Op’DD at this time because she has developed pulmonary metastases. Her chemical findings are still reasonably well controlled but she has been on the drug for only a brief period of time.
DR. DAVID W. RICHARDSON (Richmond): In the other adrenal, was there any evi- dence of bilateral adenomatous hyperplasia or was it atrophic and suppressed? A more general question : do you have any thoughts about what the stimulus to bilateral adrenocortical hyperplasia might be?
DR. SALASSA: No, I do not have any idea. I think those of you who are familiar with the field realize that data such as this, when you are talking about adenomatous hyperplasia versus single nodules, has a serious flaw in it. If the surgeon finds one good sized adenoma, he takes out that adrenal and feels the other adrenal and unless there is a palpable tumor there he doesn’t remove or do even a hemisection of the other adrenal so we really don’t know if there are small nodules in the other adrenal. So I think the whole question of whether there is a spectrum of disease similar to that seen in parathyroid disorders where one sees changes from hyperplasia, to adenomatous hyperplasia, to large autonomous functioning tumors remains to be settled. The patients we have operated with malignant tumors had no atrophy of the other ad- renal.
DR. JACQUES GENEST (Montreal) : Dr. Salassa, I was told by Dr. Franz Gross at the International Congress on Steroids in Mexico this last September that Dr. Detlev Ganten, one of our fellows who had done beautiful work on tissue isorenins had re-
cently demonstrated a very large increase in adrenal isorenin content in adenomas taken from patients with primary aldosteronism, in contrast to the non-adenomatous adrenal tissue. This increase in isorenin content of adrenal adenomas occurred despite the fact that plasma renin activity is totally suppressed in such patients. I wonder if in your cases with primary aldosteronism secondary to carcinoma if it might not be interesting to measure the isorenins in the primary adrenal tumor as well as in the metastases. I think I heard you mentioning that one of your patients had a normal or slightly elevated plasma renin activity at one point in the course of his disease.
DR. SALASSA: Well, it would be interesting, but we certainly have no data on that and I do not know of any. I didn’t mention it but the plasma renin activity in the six cases excluding the one who came in quite ill and obviously had many factors that would tend to increase his plasma renin activity, were not really significantly different than in the patients with benign adenomas.
DR. BELTON A. BURROWS (Boston): I noted that you have little difficulty in classi- fying two of the six cases as being malignant and I wonder whether the definition of malignancy depended on metastases in the other four or was it based on histological features. As I was listening to your presentation it seemed apparent that two at least of the patients had metastases as one indication of malignancy. Have you had any experience with adrenal scanning with cholesterol? It has been reported as being of value in distinguishing nodules from hyperplasia and possibly also malignant tu- mors of the adrenal.
DR. SALASSA : In response to your first question, at least the pathologists in Roches- ter have great difficulty in all the endocrine glands distinguishing on histologic ex- amination alone a large benign endocrine tumor from a low grade carcinoma. This is true of islet cell tumors of the pancreas that produce an excess of insulin. It is also true of pituitary tumors and it is certainly true of adrenocortical tumors. In three of the six patients, two have died and one has metastases so there is no question about these three. The other patient who had the 1,050 gm. tumor had gross hemorrhage in the gland and enough features of malignancy that everybody agreed that it was malig- nant. The 45 gm. and 60 gm. tumors had some features of malignancy so that a senior pathologist, Dr. Malcolm Dockerty, called them malignant. Dr. Carney who subse- quently reviewed all of the cases did not quite agree. I don’t think this is too surpris- ing. We have had a number of patients who had Cushing’s syndrome and had large tumors and the pathologist said that they were benign. Yet a number of these patients subsequently returned with metastases. So I don’t think that the pathologist on histologic grounds alone can consistently separate a large benign functioning endo- crine tumor from a low grade functioning carcinoma. In answer to the second ques- tion, we do use I131 cholesterol in an effort to locate the tumor by external scanning once the chemical diagnosis has been made. We usually do an excretory urogram. If this is negative, we then do a bolus nephrotomogram. If this is also negative, we used to then consider arteriography or retrograde adrenal venography. We now do an I131 cholesterol scan if the excretory urogram and bolus nephrotomograms are negative. I do not think we have had enough experience to really make a definite statement but it is our feeling that this has been a very helpful procedure to date. I don’t think we have had a tumor less than approximately 1.5 cm. that has picked up enough iodo- cholesterol so that we could localize the tumor and I would like to point out that in our hands the bolus nephrotomogram is about 75-80 per cent accurate if the tumor is 2.5 cm. or more in diameter. I think the reason some of these small tumors pick up the iotocholesterol is that they are full of fat. We have had several lipid ladden nonfunc- tioning benign adrenal tumors (nonfunctioning in every way that we could test for)
that also picked up I131 cholesterol. So I think that the cholesterol uptake by tumors of the adrenal may in some way be related to the total fat content of the tumor. These malignant tumors we have discussed today were rather “meaty” tumors. They were not soft and fatty and they did not pick up I131 cholesterol.
DR. NICHOLAS P. CHRISTY (New York) : Laragh and I did a study in 1961 (New Eng. J. Med. 265: 1083, 1961) in which we thought that the usual endocrine basis for hy- pokalemia in the general run of patients with Cushing’s syndrome was not hyper- aldosteronism, but rather a gross excess of cortisol. We were able to define quite a clear correlation between the depression of the serum potassium value and the eleva- tion of the serum bicarbonate on one hand and the magnitude of either plasma or urinary cortisol level on the other. Do I understand from one of your early slides that hypokalemia in most of Cushing’s patients was related to aldosterone levels?
DR. SALASSA: Oh, no. That slide did not pertain to patients with Cushing’s syn- drome and in our experience patients with Cushing’s generally have low urinary aldosterone. I agree with you on this point but I think there is one group of patients with Cushing’s syndrome that may differ a little and that is the ectopic ACTH syn- drome which Dr. Liddle could discuss better than I. But I think there is some evidence in the literature that under very intense stimulation the adrenals of some of these patients do secrete quite a bit of deoxycorticosterone. How much of the hypokalemia and hypertension in this syndrome is due to the very, very high levels of cortisol, which is of course a relatively weak mineralocorticoid, and how much is due to in- creased secretion of deoxycorticosterone, which is a relatively strong mineralocorti- coid, I do not know.