routine tests of haemostatic function are invariably nor- mal: No deficiency of platelet factor as evidenced by the normal thromboplastin generation tests was detected in either of the above patients.
In both patients the platelet count returned to normal and bleeding ceased after treatment with radioactive phosphorus.
I thank Professor J. C. Goligher, Dr. R. N. Tattersall, and Mr. G. H. Wooler for permission to study patients under their care, and Dr. W. Goldie for investigation of the coagulation factors.
REFERENCES
Biggs, R., and Douglas, A. S. (1953). J. clin. Path., 6, 23. Brugsch, H. (1933). Folia haemat. (Lpz.), 49, 454. Dameshek, W. (1951). Blood, 6, 372.
Epstein, E., and Goedel, A. (1934). Virchows Arch. path. Anat., 292, 233.
Fanger, H., Cella, L. J., and Litchman, H. (1954). New Engl. J. Med., 250, 456.
Hardisty, R. M., and Wolff, H. H. (1955). Brit. J. Haemat., 1, 390. Mortensen, O. (1948). Acta med. scand., 129, 547.
Uotila, U. (1938). Ibid., 95, 136.
CUSHING’S SYNDROME IN CHILDHOOD REPORT OF CASE OF ADRENOCORTICAL CAR- CINOMA WITH EXCESSIVE ALDOSTERONE PRODUCTION
BY
W. P. U. JACKSON, M.D., M.R.C.P., D.C.H. B. ZILBERG, M.B., M.R.C.P.Ed., D.C.H. B. LEWIS, Ph.D., M.B., A.R.I.C. AND D. McKENZIE, M.Med.(Path.)
From the Endocrine Clinic and Laboratories of the Depart- ment of Medicine ; and the Departments of Paediatrics and Pathology, Groote Schuur Hospital and the University of Capetown, South Africa
Adrenocortical overactivity in childhood may produce two distinct syndromes. The adrenogenital syndrome is caused by excessive production of androgens, and may manifest itself in three ways. If it starts in intrauterine life in a female, pseudohermaphroditism results ; post- natally in a girl it produces virilism ; in a boy the ” pocket Hercules ” variety of isosexual precocity results. An excess of glucocorticoids, on the other hand, gives rise to Cushing’s syndrome, without masculinization.
Almost all cases of Cushing’s syndrome in childhood are caused by adrenocortical carcinoma. Wilkins (1948) collected 70 cases of cortical carcinoma from the litera- ture, and in his book (Wilkins, 1950) he reviewed 26 cases of Cushing’s syndrome, all under the age of 10 years. Twenty-two of these were in girls ; as in the adult, Cushing’s syndrome in childhood is more common in females. A few other cases have been reported since then, and Guin and Gilbert (1956) produced a further review. The two cases of Cushing’s syndrome not asso- ciated with carcinoma were reported by Chute et al. (1949) (bilateral nodular hyperplasia) and Powell et al. (1955) (bilateral cortical adenomas).
Functioning adenomas of the adrenal cortex at any age produce either pure Cushing’s syndrome or a pure adrenogenital syndrome. Overproduction of the other two groups of adrenocortical hormones, the oestrogens and mineralocorticoids, appears to be of little import- ance in childhood. Oestrogen excess, leading to pre- cocious pubertal development of female type, is ex-
tremely rare, while primary aldosteronism has not, to our knowledge, been described in childhood.
Unlike adrenal adenomas, carcinomas are biochemi- cally pluripotent and produce excessive quantities of androgens, glucocorticoids, and also oestrogens (where these have been measured). In the case described below aldosterone was also found in great quantities in the urine. Consequently it is logical that the clinical picture may also be a mixed one in cases of carcinoma, being composed of features of Cushing’s and the adrenogenital syndrome :
ADRENOCORTICAL CARCINOMA
☒ Androgens + +
Glucocorticoids + +
Oestrogens ++
Mineralocorticoids + +
Cushing’s syndrome
Adrenogenital syndrome
Rarely clinical effects.
? Clinical effects,
Protein catabolism
Protein anabolism
Obesity. Weak- ness
Increased strength and muscles.
(Antagonistic effects)
Cessation of growth. Osteoporosis.
Rapid growth.
Prevention of menarche. Plethora, etc.
Precocious sexual development of male type.
Since virtually all cases of Cushing’s syndrome in childhood are caused by carcinoma one may expect to find evidence of the adrenogenital syndrome in all of them. On the other hand, only some adrenogenital cases (those with carcinoma) will show features of Cushing’s syndrome. Now, in some ways the androgens and the glucocorticoids are mutually antagonistic, so that the resultant syndrome of excess of both may be vari- able. Thus androgens promote rapid skeletal growth and maturation, whereas cortisol (the chief glucocorti- coid) abruptly terminates these. Androgens increase muscle bulk and power ; cortisol diminishes these, pro- ducing fat from muscle and so increasing weakness. Cortisol actually opposes anabolism, leading to porosis of bone and gluconeogenesis with raising of blood sugar. Androgens tend to counteract these by promoting pro- tein anabolism. In the case presented below the two opposing forces seemed so well titrated against each other that bone and tooth age and density, muscular con- figuration and power, and blood-sugar level did not appear to depart from the normal.
Case Report
A coloured girl was admitted to Groote Schuur Hospital at the age of 24 years. Since the age of 1 year there had been an abnormally rapid gain in weight. At this age pubic and axillary hair first appeared. Three weeks before admis- sion a facial eruption was noticed. There was no history of menstruation.
Physical examination revealed an obese child, with a Cushing-like facies and signs of virilism (Fig. 1). The obesity included limbs as well as trunk. Her height was 314 in. (80 cm.) and weight 39 1b. (17.7 kg.). She had a plethoric appearance and a well-marked ” buffalo-neck.” There was an eruption on the face, which resembled adenoma sebaceum. Axillary and pubic hair were present and the clitoris was enlarged. Strong body odour was obvious. There was no true breast development. Her B.P. was 130/90 ; Hb, 15 g./ 100 ml. ; P.C.V., 45%. The specific gravity of the urine was 1015. E.C.G. was normal. Blood chemistry : serum Na,
133 mEq/1. ; serum K, 5.5 mEq/1. ; serum CI, 102 mEq/1. ; CO,combining power, 57 vols.% ; blood urea, 32 mg./ 100 ml. ; serum alkaline phosphatase, 5.3 units (Bodansky- Shinowara) ; serum albumin, 4.19 g./100 ml. ; serum globulin, 2.89 g./100 ml. ; thymol turbidity, 6.3 units ; fasting blood sugar, 88 mg./100 ml .; urine concentration test, S.G. rose to 1025.
Radiographs of the long bones showed normal bone age and no osteoporosis. An intravenous pyelogram revealed normally functioning kidneys. The left kidney was depressed and a mass was visible above this kidney. Chest x-ray examination revealed no abnormality.
1a
1b
1c
1d
The 24-hour urinary output of 17-ketosteroids was 39 mg., that of 17-ketogenic steroids (Norymberski) 11.6 mg., and of aldosterone 75 µg. (expected normals for these three steroids are, respectively, under 2 mg., under 3 mg., and under 10 µg. per 24 hours). .
Urinary sodium and potassium studies were made over six consecutive days. During this time basal readings were obtained and the effects of a potassium load and of pred- nisone administration were observed. (See section on steroids below.)
The combination of features of Cushing’s syndrome and virilism at this age strongly suggested adrenocortical car- cinoma, which was supported by high urinary steroid output. Pyelography indicated that the tumour was on the left side.
Course and Treatment
On March 24, 1957, 50 mg. of cortisone was given by intramuscular injection. On the morning of the 25th this was repeated, and that afternoon left-sided adrenalectomy was performed (Professor J. H. Louw), while the patient received an intravenous infusion of N/2 saline with 50 mg. of soluble cortisol (hydrocortisone). Twelve hours post- operatively the blood pressure dropped from 140/90 to 100/70. Fifty units of soluble cortisol was given intravenously and 50 mg. intramuscularly. 50 mg. of cortisone was then given intramuscularly twice daily for three days. This was re- placed by oral cortisone in decreasing dosage until it was withdrawn after five days, when corticotrophin Z, 25 units, was injected intramuscularly on two occasions. Intravenous infusions were maintained for two days after operation. The subsequent post-operative course was uneventful.
Five weeks after the operation the patient had lost 3 1b. (1.4 kg.). She was no longer as plethoric as before ; the facial eruption had diminished, and the body odour was less obvious. Blood pressure had dropped to 105/70. Twelve weeks after the operation her appearance had changed dramatically (Fig. 1). She had lost 4 1b. (1.8 kg.) more, her facial skin was quite clear, the buffalo-hump had dis- appeared, the body odour had disappeared, and the clitoris was smaller. Hirsuties was still considerable. The 24-hour urinary output of 17-ketosteroids had fallen to 1 mg.
Seven months after the operation the clitoris was of nor- mal size and the pubic hair had almost disappeared.
Endocrine Studies ‘
In addition to estimation of hormone output levels in basal state, the effect upon these of potassium loading and of prednisone was investigated. In primary hyperaldosteronism (due to adrenal adenoma) the aldosterone output may become demonstrably abnormal only during periods of high potassium intake (Eales and Linder, 1956). The tumour in such instances resembles the normal adrenal glomerulosa in its secretory response to hyperpotassaemia. . In the present case we therefore observed the effect of an extra 100 mEq of potassium in the diet.
Secondly, the dependence of the adrenal tumour on the pituitary was studied by following the plasma cortisol and urinary steroid hormone excretion during the administration of 41-cortisone (prednisone) in doses of 5 mg. twice a day. Prednisone is one of the most potent suppressors of the secretion of corticotrophin by the pituitary.
Estimations of intake and urinary output of sodium and potassium were undertaken at the same time, to ascertain whether the cortical steroids were producing excessive reten- tion of sodium or loss of potassium.
The urinary 17-ketogenic steroids and 17-ketosteroids were determined by the method of Norymberski, Stubbs, and West (1953), plasma cortisol by that of Lewis (1957), and urinary aldosterone by the physico-chemical procedure of Neher and Wettstein (1956). The ketogenic steroids are a measure of the metabolites of cortisol predominantly, but the excre- tory product of 17-hydroxyprogesterone also contributes to these urinary steroids. 17-Hydroxyprogesterone is andro-
genic, but is probably not secreted by the normal adrenal ; it is an intermediate product in the course of the biosynthesis of glucocorticoids. Further confirmation of the nature of the substance measured by the Neher and Wettstein pro- cedure was provided by chromatography of the acetylated urinary extract in benzene/formamide, whereby the appro- priate quantity of a substance with the Rf value of aldo- sterone diacetate was demonstrated. The material also migrated with aldosterone in the system isooctane/tert- butanol-water (Eberlein and Bongiovanni, 1955).
Results
The steroid determination (Table I) showed a striking excess of aldosterone, 17-ketosteroids, and 17-ketogenic steroids in the urine and of cortisol in the plasma, represent-
| % | Urinary 17- ketosteroids mg./24 Hours | mg. /24 Hours Urinary 17- ketogenic Steroids | Plasma Cortisol µg./100 ml. | Urinary Aldosterone ug./24 Hours |
|---|---|---|---|---|
| March 3: Normal | ||||
| diet | 39 | 11-6 | ||
| , 5: | - | - | 32 | 75 |
| 6: Potassium | ||||
| load | 48 | 5-8 | ||
| " 7: | - | - | 37 | 75 |
| 8: Predni- | ||||
| sone | 43 | 6-9 | - | |
| 9: | 37 | 4.0 | 26 | - |
| 25 ", | Adrenalec- tomy | |||
| April 7 .. | - | - | 10 | |
| 15 .. | 2.8 | - | ||
| May 10 .. | - | 4 | 2 | |
| 11 | 1.0 | - | - | - |
ing hypersecretion of mineralocorticoid, androgen, and gluco- corticoid. Fractionation of the 17-ketosteroids showed an abnormal amount of dehydroepiandrosterone. This is the main constituent of the ” ß-fraction ” as segregated by digitonin precipitation, and is believed to be strongly sugges- tive of adrenocortical carcinoma.
Potassium loading (100 mEq in addition to normal diet) failed to augment the massive aldosterone output. Pred- nisone (given for only 48 hours) showed a possible suppress- ing effect on cortisol and 17-ketosteroid secretion on the second day, but the results were not conclusive and this suppression, if present, was not great.
Sodium and potassium ” balance ” data are presented in Table II. Urine collections were difficult, and were analysed only on those days when collections were virtually complete.
| Sodium (mEq/24 Hours) | Potassium (mEq/24 Hours) | |||
|---|---|---|---|---|
| Intake | Urine | Intake | Urine | |
| March 5: Normal diet | 45.4 | 22.3 | 63.9 | 32-3 |
| 6: High K intake . . | 43-8 | 41-0 | 149.9 | 74.9 |
| ,, 8: High K intake and prednisone | ||||
| 47-0 | 36-4 | 157.2 | 78-8 | |
| . 9: , | 45-1 | 33.0 | 152.9 | 81.4 |
We observed no evidence of any considerable retention of sodium, and certainly no excessive loss of potassium. The high potassium intake appeared to augment the urinary sodium. The prednisone had no effect on the electrolyte balance.
The gratifying clinical response to adrenalectomy was mirrored in the hormone studies, which declined to normal levels (and have remained so during a six-months follow-up).
The Adrenocortical Tumour
The specimen (Fig. 2) consisted of a triangular mass of tissue weighing 36 g. and measuring 5 cm. along each border and 3 cm. in thickness. The outer surface was smooth and encapsulated. A small breach at one point was apparently due to operative trauma. At one pole there was a project- ing tag of orange-yellow tissue measuring 1.5 by 2 by 1 cm.
On cut section the greater portion was lobulated and bright orange- yellow, a smaller portion appearing more fleshy and brown.
Sections stained with haematoxylin and eosin showed a highly anaplastic carcinoma (Fig. 3). The majority of the cells of the tumour were ex- tremely bizarre in shape and varied considerably in size. The cyto- plasm of many cells was finely granular and vacu- olated. The nuclei INCHES 1l 21 showed many ir- FIG. 2 .- Left adrenal gland after removal. regularities and mitotic figures were numerous. The tumour cells were divided haphazardly into groups by a fairly extensive net- work of thin-walled vessels. There was no haemorrhage into the tumour, and no obvious vascular infiltration was noted.
Closely applied to the tumour on its external surface was a well-marked fibrous tissue capsule. On the outer surface of this, and separated by it from the tumour, was a tag of compressed normal-looking adrenal gland tissue.
A frozen section showed varying amounts of fat in dif- ferent areas. In some areas cells were loaded with fat ; in others it was evident as fine droplets in the cytoplasm.
Ponceau-fuchsin staining of a section fixed in formol- saline showed the cytoplasm of the tumour cells to be loaded with vivid red fuchsinophil granules which were not evi- dent in the attached portion of normal adrenal.
Discussion
Diagnosis .- In this child the clinical features of Cushing’s syndrome and virilism suggested the diagnosis of adreno- cortical carcinoma. The very high urinary output of 17- ketosteroids provided confirmatory evidence. An interesting
feature here was the apparent metabolic “balance ” produced by the concomitant hypersecretion of glucocorti- coids, mineralocorticoids and androgens. Thus there was no disturbance in muscular strength, bone growth, or electro- lyte balance. The marked body odour probably was due to excessive apocrine activity. The tumour was lateralized by intravenous pyelography combined with tomography. Pre- sacral air insufflation was rendered unnecessary by this manœuvre. It is noteworthy that the length of this history (18 months) was not incompatible with a localized carcinoma in this case.
Steroid Findings .- Despite the probable zonation of function within the normal adrenal cortex (Chute et al., 1949 ; Landing, 1955), the tumour, a strikingly undifferenti- ated carcinoma, possessed the ability to synthesize all the three major groups of adrenal hormones. While the simul- taneous hypersecretion of glucocorticoid and androgen by adrenal tumours is commonplace, and that of glucocorticoid and aldosterone has been described, the present case offers a remarkable example of pluripotency of an adrenal tumour. The increase in aldosterone excretion was notable not only for its magnitude but also for the absence of its expected effect on electrolyte excretion or serum potas- sium level. The opposing effect of androgens in promoting potassium retention may be relevant in this connexion. In. common with some cases of Conn’s syndrome of primary aldosteronism, sodium metabolism was not grossly dis- turbed ; arterial hypertension was, however, present. Possible explanations for lack of electrolyte disturbances include the simultaneous hypersecretion of an unknown sodium-diuretic hormone ; alternatively, some combination of known hormones may possess sodium-diuretic activity, or conceivably the renal tubular epithelium may lose its responsiveness to aldosterone.
Treatment .- When there is no evidence of secondary deposits, adrenalectomy is indicated in cases of adreno- cortical carcinoma. When there is doubt about which adrenal is involved, the transperitoneal approach is favoured. Adrenalectomy may be followed by a period of hypo- adrenalism, because the opposite adrenal is usually atrophic. To prevent this, cortisone is usually given pre-operatively and continued during the operation and in the immediate post- operative period. When cortisone is withdrawn, corticotrophin may be given in an effort to stimulate the remaining adrenal gland. In our patient there was a precipitous drop in the blood pressure 12 hours post-operatively. This was believed to be an indication for intravenous cortisol. It sug- gested that the remaining adrenal was atrophic and not also carcinomatous. The change in the appearance of the patient 12 weeks after the operation was very gratifying. A striking feature was the disappearance of the strong body odour. Though androgen overactivity as judged by the 17- ketosteroid excretion was now controlled, the hirsuties and clitoral enlargement did not disappear until a further four months had passed. These signs of virilism usually take a long time to regress and may in some instances be per- manent, in which case clitoridectomy may be necessary.
Prognosis .- If the offending adrenal is removed before metastases have developed, the chances of complete cure appear to be good. Unfortunately, in many cases of Cushing’s syndrome in childhood the progress is rapid and adrenalectomy useless. Our patient seems to be fortunate, and the probability of cure must be considered high, since steroid excretion has returned to a normal level.
Summary
The relationship between Cushing’s syndrome in child- hood and adrenocortical carcinoma is briefly considered. A “mixed picture ” of Cushing’s and adrenogenital syndromes is characteristic, and is illustrated by a reported case. In this patient an excessive production of glucocorticoids, androgen, and mineralocorticoids (aldosterone) was demonstrated. In some ways the
usual clinical effects of each of these steroids were annulled by the opposing effects of the others.
The offending adrenocortical carcinoma was removed. Its histological appearance is described. The operation was followed by a remarkable improvement, with return to normal in the child’s appearance and her hormone output.
We thank Professors J. F. Brock, G. C. Linder, F. Forman, J. H. Louw, and F. J. Ford for their continued help and interest. The patient was under the care of the latter two professors while in hospital. Dr. T. Sacks took the photomicrograph. The costs were covered by grants from the Council for Scientific and Indus- trial Research of South Africa and the Staff Research Fund of the University of Capetown.
REFERENCES
Chute, A. L., Robinson, G. C., and Donohue, W. L. (1949). J. Pediat., 34. 20.
Eales, L., and Linder, G. C. (1956). Quart. J. Med., 25, 539. Eberlein, W. R., and Bongiovanni, A. M. (1955). Arch. Biochem., 59, 90. Guin, G. H., and Gilbert, E. F. (1956). Amer. J. Dis. Child., 92, 297. Landing, B. M. (1955). Ciba Foundation Colloquia on Endocrinology, 8, 52. Lewis, B. (1957). J. clin. Path., 10, 148.
Neher, R., and Wettstein, A. (1956). J. clin. Invest., 35, 800.
Norymberski, J. K., Stubbs, R. D., and West, H. F. (1953). Lancet. 1, 1276.
Powell, L. W., Newman S., and Hooker, J. W. (1955). Amer. J. Dis. Child., 90, 417.
Wilkins, L. (1948). J. clin. Endocr., 8, 111.
- (1950). The Diagnosis and Treatment of Endocrine Disorders in Child- hood and Adolescence. Thomas, Springfield.
MULTIPLE EPIPHYSIAL DYSPLASIA BY HERBERT BARRIE, M.D. CEDRIC CARTER, B.M., M.R.C.P .* AND JOHN SUTCLIFFE, M.B., M.R.C.P., F.F.R. Hospital for Sick Children, Great Ormond Street, London
Multiple epiphysial dysplasia is an uncommon con- dition, but is likely to be of interest to general prac- titioners on account of its marked familial features. Although Fairbank (1947) did not stress the genetic aspect in his original description, a number of affected families have since been recorded. Often a correct diagnosis in one member of a family will uncover several previously undiagnosed or misdiagnosed cases. Recogni- nition of this condition is of more than academic interest, since many patients derive benefit from appropriate physiotherapy, while harm may result from treatment based on the wrong diagnosis. To make this condition more widely known, we report three further affected families, with descriptions of the clinical, radiological, and genetic features of these and previously pub- lished case reports. The family trees are shown in Figs. 1, 2, and 3. The clinical features are summarized in the Table.
Historical Note .- In 1935 Fairbank suggested the term “epiphysial dysplasia ” for a patient with multiple irregular epiphysial ossification. To this title ” general- isata ” was added, but it was changed later to ” multi- plex ” when the epiphysial changes were found to be multiple rather than generalized. Twelve years later he published a masterly account of the clinical and radio- logical features of the disease based on a study of 20 patients, many of whom had previously been reported as atypical examples of a variety of conditions, includ- ing achondroplasia, stippling of the epiphyses (dys-
*Medical Research Council, Clinical Genetics Research Unit.