C 1988 S. Karger AG, Basel 0301-0163/88/0306-0217$2.75/0
Testosterone-Secreting Adrenocortical Tumor in a Pubertal Girl Case Report and Review of the Literature1
W. Sorgoª, D. Meyerª, K. Rodensa, J. Homokia, E. Heinzea, B. Heymerb, R. Siebenmannc
Departments of ª Pediatrics and Pathology, University of Ulm, FRG; ·Department of Pathology, Stadtspital Triemli, Zurich, Switzerland
Key Words. Testosterone . Adrenocortical tumor . Deep voice
Abstract. A girl aged 12 years and 10 months presented with deepening of the voice first noted 7 months earlier. Pubertal development was almost completed. The girl had regular monthly menses and no signs of hirsutism, clitoris enlargement or Cushing’s disease. Serum testosterone was about threefold above normal, whereas dehydroepian- drosterone was in the upper normal range. The 17-ketosteroids as well as the gas-chromatographically analyzed 5a and 5ß derivatives of testosterone from urine were slightly increased. Other serum and urinary steroids were normal. Dynamic tests of the endocrine function exhibited inconclusive results. Ultrasonography revealed no ovarian cysts. A small, left-sided adrenal mass was identified by computed axial tomographic scan and removed by surgery. There were no signs of local metastasis nor of vascular extension. The histopathological diagnosis was adrenocortical carcinoma. 5 months after surgery, the preoperatively elevated steroid levels had returned to normal.
Introduction
Adrenocortical tumors are rare in both children and adults [1], representing about 0.2% of all cancers in childhood [2] and 0.02% of all cancers in the general population [3]. Adrenocortical tumors in children seem to cluster in the first decade of life [4], whereas in adults they occur most frequently at the age of 40-50 years.
The tumor has a preponderance in females with a female:male ratio of 2.3:1 [5]. In some cases, a familial incidence with other malignancies as well as sibship occurrence have been observed [6]. Although nonsecret- ing adrenal tumors have been reported [6], most of those seen in pediatric patients are functionally active leading to Cushing’s syndrome, virilization, feminization, hy- peraldosteronism or a combination of these features [3]. In a review of 222 pediatric cases, it was found that two thirds had virilization as the predominant clinical symp- tom, and one third hypercortisolism [5]. More than half
of the virilizing adrenal tumors in three pediatric series were carcinomas [7].
The present report is the first on a pubertal girl pre- senting with deepening of the voice as the only clinical sign of androgenization. A small testosterone-secreting adrenal tumor with borderline increased excretion of urinary 17-ketosteroids (KS) was localized by computed axial tomographic (CAT) scanning and removed by sur- gery. The histologic diagnosis suggested an adrenocorti- cal carcinoma. The validity of several diagnostic proce- dures was analyzed and compared to reported data from the literature.
Materials and Methods
Total urinary 17-KS were determined by the procedure of Peter- son and Pierce [8]. Gas-chromatographic analysis of urinary ste- roids was carried out as delineated by Shackleton and Honour [9]. Standard radioimmunoassay (RIA) kits were used for the following determinations: serum testosterone, androstenedione, dehydroepi- androsterone sulfate (DHEA-S), sexual-hormone-binding globulin (SHBG), luteinizing hormone (LH) and follicle-stimulating hor- mone (FSH; bioMérieux, Charbonnières-les-Bains, France). LH and
1 Dedicated to Prof. W.M. Teller on the occasion of his 60th birthday.
| Steroid | Before surgery | After surgery basal | Normal range | ||||
|---|---|---|---|---|---|---|---|
| basal | Cortrosyn! (60 min) | dexamethasone! (2nd day) | hCG1 (4th day) | ||||
| low | high | ||||||
| Serum/plasma2 | |||||||
| Testosterone, nmol/l | 7.28 | 6.94 | 4.98 | 2.71 | 5.20 | 0.97 | 0.35-2.433 |
| Androstenedione, nmol/l | 6.29 | ND | 6.46 | 6.29 | 9.78 | 5.94 | 2.97-9.603 |
| DHEA, nmol/l | 30.34 | ND | 9.54 | 0.35 | 45.08 | 16.85 | 9.71-31.213 |
| DHEA-S µmol/l | 6.26 | ND | 5.03 | 4.76 | 6.12 | 5.44 | 3.62-10.883 |
| Cortisol, umol/l | 0.28 | 0.75 | 0.06 | 0.04 | 0.20 | 0.14-0.554 | |
| 11-Desoxycortisol, pmol/l | 5.20 | 8.90 | 2.89-8.664 | ||||
| 11-Desoxycorticosterone, nmol/l | 0.05 | 0.37 | 0.03-0.394 | ||||
| 17a-OH-progesterone, nmol/l | 1.97 | 6.66 | 1.48 | 1.30-6.663 | |||
| 21-Desoxycortisol, pmol/l | 0.35 | 1.12 | 0.06-0.434 | ||||
| 17a-OH-pregnenolone, nmol/l | 3.97 | 7.52 | 3.91 | 0.90-10.534 | |||
| Urine | |||||||
| 17-KS, umol/d | 46.50 | 30.80 | 19.40 | 24.30 | 6.90-34.703 | ||
| Free cortisol, umol/d | 0.21 | 0.09 | 0.03 | 0.16 | 0.10-0.334 | ||
ND = Not done; insufficient sample.
1 As described in Materials and Methods.
2 Levels at 9 a.m.
3 For girls from 14 to 16 years of age.
4 For adults.
FSH standards were calibrated each versus LER 907; plasma estra- diol (Immunchem., Carson, Calif.); serum cortisol, urinary free cor- tisol, aldosterone, prolactin, human chorionic gonadotropin (B- hCG), a -fetoprotein and carcinoembryonic antigen (CEA; Diagnos- tic Products, Los Angeles, Calif.); plasma renin (Institut Pasteur, Lyon, France); plasma levels such as dihydrotestosterone (DHT), 11-desoxycortisol, 11-desoxycorticosterone, DHEA, 17a-hydroxy- prenenolone, 17a-hydroxyprogesterone and 21-desoxycortisol were determined applying RIA procedures developed by Vecsei [10]. Plasma adrenocorticotropin (ACTH) was assayed according to a radioimmunological method described by Voigt et al. [11]. The assessment of adrenal function was performed by stimulation with 0.25 mg Cortrosyn/m2 by intravenous injection. Blood samples for steroid measurements were collected before and 60 min following application. Suppression of adrenal steroid secretion (plasma and urine) was investigated following oral administration of dexametha- sone (‘low dose’ = 0.5 mg every 6 h for 2 days, and ‘high dose’ = 2 mg every 6 h). To determine the presumptive source of androgen excess (i.e. gonadal or adrenal), an hCG stimulation test was carried out by intramuscular injection of 4,000 IU once daily for 3 consecutive days. LH-releasing hormone (LH-RH) was given as an intravenous bolus of 100 µg and blood samples for FSH and LH measurements were drawn at 0, 45 and 60 min.
Height and weight were related to the Standard Growth Curves of the Kinderspital Zurich (not yet published).
Case Report
The girl aged 12 years and 10 months was referred to the pediat- ric outpatient clinic for evaluation of a deep voice first noted 7 months before. It was because of the deepening of the voice that she had become unhappy and withdrawn from her schoolfriends. Any administration of an anabolic steroid was denied. There was no family history of malignancy. The girl was the single child of healthy parents and had never been seriously ill before. On physical exami- nation, she had a deep voice but no clinical signs of hirsutism, cli- toromegaly, acne or Cushing’s syndrome. Pubic hair and breast development corresponded to Tanner stage IV and V, respectively [12]. The girl had menarche at the age of 12 years and 7 months and regular monthly menses since then. Muscular development was not remarkable, Abdominal and gynecologic examination revealed no palpable masses. There was no goiter nor any other internal, neuro- logic or cutaneous abnormalities [6]. The phoniatric examination revealed no abnormalities of the vocal tract.
Bone age [13] was 15 years and was accelerated by more than 2 years compared to chronologic age. Height was 150 cm (10 .- 25. percentile, P), weight 47.3 kg (50 .- 75. P), height prediction 152 cm [14], target height 162 ± 8.5 cm [15]. Previous growth data were not available. Blood pressure was 120/80 mm Hg.
Laboratory evaluation pre- and postoperatively revealed normal values for blood count, urinalysis, serum sodium, potassium, cal-
Downloaded by:
B
5 0 +267
CONTRAST
R
L
L
A
ST
¥
S
Vy 308 ML +117 -33
1
AC
PC
T
2a
C
P
S
2b
cium, phosphorus, blood urea nitrogen, creatinine, glucose, serum aspartate aminotransferase, alanine aminotransferase, lactic dehy- drogenase and alkaline phosphatase. The following values were also within normal limits for girls from 14 to 16 years of age (patient’s bone age 15 years). If such reference data were not available, normal adult standards were applied: prolactin, 4 ug/l; estradiol, 202 pmol/l; SHBG, 31.2 nmol/l; DHT, 448 pmol/l; aldosterone, 458 pmol/l; plasma renin, 10 ng/l; ACTH, 18.7 pmol/l; ß-hCG, 5 IU/1; CEA, 1.7 ng/ml, and aj-fetoprotein, 3 ng/ml. The LH-RH stimula- tion test exhibited a normal response both for LH from 4.7 up to 25.3 IU/l and FSH from 7.2 to 14.1 IU/1.
Preoperatively determined serum testosterone was about three- fold above normal. Urinary 17-KS were slightly elevated and serum DHEA was in the upper normal range (table 1). With the exception of the 5a and 5ß derivatives of testosterone which were increased, ste- roid profiling in urine including pregnanetriol, pregnanetriolone, tetrahydrodesoxycortisol (THS), tetrahydrodesoxycorticosterone (THDOC), pregnenediol and pregnenetriol was normal. Serum corti- sol and its free fraction in urine were also normal. Cortrosyn stimu- lation revealed normal responses of the adrenal plasma steroids and confirmed the urinary analysis which had already excluded an enzy- matic disorder of steroid biosynthesis as the cause of increased andro- gen excretion. Serum testosterone was unchanged after Cortrosyn.
Unstimulated serum cortisol was suppressed by high-dose dexameth- asone by about 80%, and DHEA decreased to 0.35 nmol/l while androstenedione did not change. The 17-KS level declined by more than 50%. However, serum testosterone and the gas-chromatograph- ically analyzed testosterone metabolites in urine were not appro- priately suppressed. Even after high-dose dexamethasone the testos- terone level did not reach the normal range. Serum testosterone could not be stimulated by hCG provocation, whereas androstenedione and DHEA increased both by about 50% above the basal levels.
By CAT scanning a left-sided adrenal mass was visualized mea- suring 1 X 2 cm (fig. 1). The right adrenal, both ovaries and other parts of the pelvis were normal. It was not possible to demonstrate the left-sided adrenal mass by real-time ultrasonography neither before nor after abdominal CAT scan. There were no ovarian cysts. We did not carry out either a scintiscanning of the adrenal or a catheterization of the adrenal and ovarian veins. An X-ray film of the chest disclosed no abnormalities.
Laparotomy was performed through a transabdominal ap- proach. The left adrenal showed a well-encapsulated mass of the reported size. The entire gland was removed. The right adrenal, liver and kidneys appeared normal to inspection and palpation.
Microscopic examination revealed an adrenocortical tumor which was completely encapsulated (fig. 2a). No extension of the
Downloaded by:
a
b
tumor into the capsule or into the surrounding blood vessels was found. The tumor consisted of two cell types: small compact cells and large spongiocytes (fig. 2b). There were no crystalloids of Reinke that would have suggested Leydig cell origin. The tumor cells were pleomorphic and displayed polyploidism. There was no clear- cut barrier between the compact cells and the spongiocytes. The nuclei of the latter seemed to be pycnotic and their cytoplasm exhib- ited sometimes microcalcification. In both cell populations, typical and atypical mitoses were present (fig. 3). The tumor cells were poor in lipids, and the stroma was sparse. Some lipofuscin-loaded macro- phages and some inflammatory cells were scattered throughout the tumor. The residual tissue outside the tumor showed the normal adrenal architecture. Histopathological diagnosis was adrenocorti- cal carcinoma.
Five months after surgery, the patient is doing well. There is no sign of progression of androgenization. The cyclic regulation is unaf- fected. The deep voice is persisting. The patient has never had evi- dence for adrenal insufficiency. Serum testosterone and 17-KS lev- els have returned to normal (table 1).
Discussion
The differential diagnosis of androgenization in a girl includes congenital adrenal hyperplasia, Cushing’s syn- drome caused by excess ACTH, the polycystic ovary syn- drome (PCO), adrenal adenoma or carcinoma, and ovar- ian tumors such as arrhenoblastomas, hilus cell tumors and gynandroblastomas [1]. Arrhenoblastomas and PCO are the most common ovarian cause of virilization in younger patients, while in postmenopausal women hilus cell tumors are more frequent [16].
Androgenization with elevated serum testosterone levels, normal gonadotropins and normal urinary 17-KS is usually caused by an ovarian tumor, but may also be observed in testosterone-secreting adrenal adenomas [17]. Textbook opinion holds that DHEA is commonly elevated in adrenal tumors, however, the failure to find a marked elevation of serum DHEA should not be consid- ered a deterrent when an adrenal tumor is suspected on other reasons.
In our patient the hormonal evaluation (table 1) excluded some so-called mild or late-onset forms of con- genital adrenal hyperplasia such as 21- and 11ß-hydrox- ylase or 3ß-dehydrogenase deficiencies. Cushing’s syn- drome, PCO and the ingestion of anabolic steroids could also be ruled out.
In view of the rarity of primarily testosterone-secret- ing adrenal tumors in childhood and adolescence it is not surprising that most cases reported in the literature deal with adult patients. As far as we know, 15 females with exclusively testosterone-secreting adrenal tumors and normal 17-KS excretion had been recently reviewed from the literature [18, 19]. The youngest patient was 1 year and 8 months, 2 others below 20, and the oldest 76 years of age. If one considers cases in which elevated serum testosterone levels were associated with a slightly increased urinary 17-KS fraction (up to 70 umol/d), another 5 patients (including the current case) may be added [1, 7, 17, 20]. In the synopsis of table 2, the patients are grouped according to their endocrine find- ings. There seems to be no relation between the endo- crine findings and the histopathology in these patients, although it is interesting that the 2 cases with adrenocor- tical carcinoma belonged to group II and were below 20 years of age. 11 of the aforementioned 15 patients ini- tially underwent ovarian surgery. Misdiagnosis was caused partly because the adrenals had not been visual- ized before surgery and partly because the hormonal results following stimulation or suppression tests were misleading. In our patient, the dynamic tests with Cor- trosyn and dexamethasone exhibited normal adrenal function with suppression of DHEA and urinary 17-KS excretion (table 1). These findings were initially believed to contradict the presence of an adrenal tumor. How- ever, adrenal tumors were also said to be responsive to exogenous steroid administration and ACTH stimula- tion [6]. Ovarian tumors might also secrete testosterone in response to Cortrosyn stimulation [18]. Since in our case serum DHEA was preoperatively in the upper nor- mal range whereas androstenedione was quite normal, it could be suggested that the conversion by the 3ß-dehy-
Downloaded by:
| Patient No. | Reference | Age years | Plasma testosterone nmol/l | Urinary 17-KS umol/d | Histopathology | |
|---|---|---|---|---|---|---|
| Group I | ||||||
| 1 | 32 | 1.66 | 149.10 | <0.52 | normal | ACA; RC? |
| 2 | 18 | 15 | 12.14 | normal | ACA; RC neg. | |
| 3 | 33 | 16 | 16.37 | normal | ACA; RC? | |
| 4 | 34 | 26 | 26.01 | normal | AGN; RC? | |
| 5 | 24 | 38 | 19.70 | normal | AGN/ACA; RC neg. | |
| 6 | 26 | 49 | 31.59 | normal | ACA; RC neg | |
| 7 | 19 | 49 | 48.82 | normal | ACA; RC pos. | |
| 8 | 23 | 50 | 33.63 | normal | ACA; RC? | |
| 9 | 22 | 58 | 13.90 | normal | AGN; RC pos. | |
| 10 | 16 | 60 | 27.67 | normal | ACA; RC pos. | |
| 11 | 35 | 60 | 22.54 | normal | ACA; RC neg. | |
| 12 | 25 | 61 | 19.76 | normal | ACA; RC? | |
| 13 | 36 | 62 | >17.34 | normal | ACA; RC neg. | |
| 14 | 37 | 63 | 9.88 | normal | ACA; RC neg. | |
| 15 | 38 | 76 | 31.66 | normal | ACA; RC neg. | |
| Group II | ||||||
| 1 | 1 | 0.91 | 11.62 | <0.35 | 15.60 <3.47 | ACC; RC? |
| 2 | 7 | 3 | 4.72 | <0.35 | 10.75 <6.93 | ACA; RC? |
| 3 | present case | 12.83 | 7.28 | <2.43 | 46.50 <34.67 | ACC; RC neg. |
| 4 | 20 | 22 | 11.10 <2.43 | 60.68 <52.01 | ACA; RC? | |
| 5 | 17 | 51 | 29.47 | 69.35 <55.48 | ACA; RC? | |
Normal ranges of plasma testosterone and urinary 17-KS for adult females are <3.48 nmol/l and 17.35-52.10 umol/d, respectively unless otherwise noted. Group I = Adrenal tumors secreting only testosterone; group II = adrenal tumors secreting testosterone plus low levels of 17-KS; ACA = adrenocortical adenoma; AGN = adrenal ganglioneuroma; ACC = adrenocortical carcinoma; RC = Reinke crystal- loids.
drogenase in the tumor tissue had been impaired [21]. Another explanation for this finding would imply a simultaneously increased peripheral conversion from androstenedione to testosterone [6, 20].
As in our case, serum testosterone did not change in 5 patients during dexamethasone suppression, but de- creased adequately in 3 others reported in the literature [18]. Cortrosyn did neither stimulate serum testosterone in 4 cases [22-24], whereas in 2 others there was a mod- erate increase [25]. Testosterone could be stimulated by hCG administration in 5 patients [26], but not in ours and not in the case published by Schteingart et al. [20]. To our knowledge, our patient is the first in whom hCG provocation had stimulated DHEA and androstenedione secretion. However, we must admit that the source of the surge in DHEA and androstenedione production does not necessarily concern the removed adrenal, but may also be related to ovarian tissue. A similar finding led to the tentative diagnosis of an ovarian hilus cell tumor in a 60-year-old postmenopausal woman [16].
The role of testosterone in the feedback regulation of FSH and LH secretion is difficult to understand. In our case, LH-RH stimulation resulted in normal LH and FSH responses. From this and the clinical fact that the girl had normal monthly menses, it can be confirmed that increased testosterone levels do not necessarily in- fluence the secretion of gonadotropins [27].
Furthermore, one wonders how testosterone as the most potent endogenous androgen should be responsible for the deepening of the voice and probably also for the bone age acceleration, without having effected other con- comitant signs of androgenization. In contrast, 5 other cases below 20 years of age (table 2) presented with vir- ilization and acne at first examination [1, 7, 18, 32, 33]. Additional findings were a deep voice in 4 [1, 18, 32, 33], amenorrhea in 2 pubertal [18, 33] and growth accelera- tion in 2 prepubertal patients [7, 32]. Therefore, it must be suggested that the sensitivity and response of andro- gen-dependent tissues to testosterone differ. It may also be possible that the unusual onset of androgenization is
caused by a temporal dissociation of the testosterone effect.
CAT scanning is suggested to be by far the most reli- able imaging procedure to detect adrenal tumors [7, 18]. However, in 3 patients with primary adrenal carcinoma in whom real-time and duplex sonography were simulta- neously employed, the sonography proved to be the superior technique for identifiying adrenal masses [28]. In our case as well as in 2 others [18], ultrasound failed to detect the tumor, whereas this procedure was also suc- cessful in the case published by Kaplowitz and Mandell [7]. CAT scan provided correct diagnosis in all 6 patients described by Kamilaris et al. [18].
In addition, adrenal and ovarian vein catheterization may also be helpful for measuring testosterone levels. However, these procedures are complicated by the risks of extravasation of contrast media and adrenal, ovarian or renal thrombosis [7, 18, 29]. Apart from this, the results of selective venous sampling may also be conflict- ing, because the adrenal and ovarian vein of the left side drain both into the renal vessel, so that the origin of the testosterone samples obtained in the effluent may re- main uncertain.
Another diagnostic procedure to localize an adrenal tumor is scintiscanning. Gross et al. [30] reported that an abnormal adrenal anatomy described by CAT scan was functionally reflected by iodocholesterol scintigraphy in 8 patients with endocrine active adrenal masses. The procedure was described to be successful in some [18, 20], but also to be unsuccessful [16] in other cases.
Since adrenal tumors are usually detectable by CAT scanning, ovarian and adrenal catheterization and scin- tiscanning are rarely necessary. This may also be valid for some of the dynamic hormonal tests which at times give controversial results [18]. In general, if an adrenal or ovarian tumor is clinically suspected, but cannot be identified by CAT scan or ultrasonography, the tumor will be very small and most probably located in the ovary (e.g. hilus cell tumor).
The histology in the present case was that of an adre- nocortical carcinoma without proof of Reinke crystal- loids which are thought to be the unique specific marker for steroid-producing Leydig cells. The failure to demon- strate Reinke crystalloids on the other hand does not necessarily exclude Leydig cell origin, because those inclusions are present in only 40% of Leydig cell tumors and 44% of ovarian hilus cell tumors [19]. Those crystal- loids were reported in only 3 testosterone-secreting adre- nal adenomas [16, 19, 22] (table 2). The prognosis of an adrenocortical carcinoma remains doubtful [7]. In the
series of Hayles et al. [5], 1 of the 12 cases reported had metastases when first seen. None of the tumors in the other patients behaved in a malignant fashion. By con- trast, in the series of Kenny et al. [4], 4 of 5 patients with adrenocortical carcinoma died within 4 years postopera- tively from widely spread metastases.
Conclusion
The present case is unusual in that a deep voice was the only clinical symptom of a small testosterone-secret- ing adrenocortical carcinoma in a pubertal girl aged 12 years and 10 months. As could be shown, the signifi- cance of the hormonal findings in such patients may not be conclusive. If a diagnosis remains unclear, i.e. endo- crine data and visualizing techniques fail to confirm the tumor, exploratory laparotomy is indicated.
In the future, it may become possible to distinguish adrenocortical adenomas and carcinomas from benign hyperplasia by demonstrating monoclonality in the former using molecular genetic methods. Such data, which have been elaborated successfully in the clarifica- tion of the origin of parathyroid tumors [31], may be very helpful with respect to a more precise histomorpho- logical diagnosis and clinical prognosis.
Acknowledgment
We are indebted to Prof. Dr. P. Vecsei (Department of Pharma- cology, University of Heidelberg, FRG) for measuring plasma DHT, 11-desoxycortisol, 11-desoxycorticosterone, DHEA, 17a-hydroxy- pregnenolone, 17a-hydroxyprogesterone and 21-desoxycortisol.
References
1 Bauman A, Bauman CG: Virilizing tumor of adrenal cortex in an infant. NY State J Med 1982;82:1070-1072.
2 Hartley HL, Birch JM, Marsden HB, et al: Adrenal cortical tumors: Epidemiological and familial aspects. Arch Dis Child 1987;62:683-689.
3 Samaan NA, Hickey RC: Adrenal cortical carcinoma. Semin Oncol 1987;14:292-296.
4 Kenny FM, Hashida Y, Askari A, et al: Virilizing tumors of the adrenal cortex. Am J Dis Child 1968;115:445-458.
5 Hayles AB, Hahn HB, Sprague RG, et al: Hormone secreting tumors of the adrenal cortex in children. Pediatrics 1966;37:19- 25.
6 Lee PDK, Winter RJ, Green OC: Virilizing adrenocortical tu- mors in childhood: Eight cases and a review of the literature. Pediatrics 1985;76:437-444.
7 Kaplowitz PB, Mandell J: Virilizing adrenal tumor in a 3 year old girl: Diagnosis by computed tomography scan and ultra- sound. Am J Dis Child 1983;137:406-407.
8 Peterson RE, Pierce CE: Determination of urinary neutral 17- ketosteroids; in Sunderman FW, Sunderman FW Jr (eds): Lipids and the Steroid Hormones in Clinical Medicine. Philadelphia, Lippincott, 1960.
9 Shackleton CHL, Honour JW: Simultaneous estimation of uri- nary steroids by semiautomated gas chromatography: Investiga- tion of neo-natal infants and children with abnormal steroid syn- thesis. Clin Chim Acta 1976;69:267-283.
10 Vecsei P: Glucocorticoids: Cortisol, cortisone, corticosterone, compound S, and their metabolites; in Jaffe BM, Behrmann HR (eds): Methods of Hormone Radioimmunoassays. New York, Academic Press, 1979.
11 Voigt KH, Fehm HL, Reck R, et al: Spontaneous and stimulated secretion of Quso extractable immunoassayable ACTH in man. Klin Wochenschr 1974;52:516-521.
12 Tanner JM: Wachstum und Reifung des Menschen. Stuttgart, Thieme, 1962.
13 Greulich WW, Pyle SI: Radiographic Atlas of Skeletal Develop- ment of the Hand and Wrist, ed 2. Stanford, Stanford University Press, 1959.
14 Bayley N, Pinneau SR: Tables for predicting adult height from skeletal age: Revised for use with the Greulich-Pyle hand stan- dards. J Pediatr 1952;40:423-441.
15 Tanner JM, Goldstein H, Whitehouse RH: Standards for chil- dren’s height at ages 2-9 years allowing for height of parents. Arch Dis Child 1970;45:755-762.
16 Trost BN, Koenig MP, Zimmermann A, et al: Virilization of a postmenopausal woman by a testosterone secreting Leydig cell type adrenal adenoma. Acta Endocrinol (Copenh) 1981;98:274- 282.
17 Werk EE, Sholiton LJ, Kalejs L: Testosterone secreting adrenal adenoma gonadotrophin control. N Engl J Med 1973;289:767- 770.
18 Kamilaris TC, DeBold CR, Manolas KJ, et al: Testosterone secreting adrenal adenoma in a peripubertal girl. JAMA 1987; 258:2558-2561.
19 Vasiloff J, Chideckel EW, Boyd CB, et al: Testosterone secreting adrenal adenoma containing crystalloids characteristic of Leydig cells. Am J Med 1985;79:772-776.
20 Schteingart DE, Woodbury MC, Tsao HS, et al: Virilizing syn- drome associated with an adrenal cortical adenoma secreting predominately testosterone. Am J Med 1979;67:140-146.
21 Agata RD, Malozowski S, Barkan A, et al: Steroid biosynthesis in human adrenal tumors. Horm Metab Res 1987;19:386-388.
22 Aguirre O, Scully RE: Testosterone secreting adrenal ganglion- euroma containing Leydig cells. Am J Surg Pathol 1983;7:699- 705.
23 Smith HC, Posen S, Clifton-Bligh P, et al: A testosterone secret- ing adrenal cortical adenoma. Aust NZ J Med 1978;8:171- 175.
24 Takahashi H, Yoshizaki K, Kato H, et al: A gonadotrophin responsive virilizing adrenal tumour identified as a mixed gan- glioneuroma and adrenocortical adenoma. Acta Endocrinol (Co- penh) 1978;89:701-709.
25 DeLange WE, Pratt JJ, Doorenbos H: A gonadotrophin respon- sive testosterone producing adrenocortical adenoma and high gonadotrophin levels in an elderly woman. Clin Endocrinol 1980;12:21-28.
26 Givens JR, Andersen RN, Wiser WL, et al: A gonadotropin responsive adrenocortical adenoma. J Clin Endocrinol Metab 1974;38:126-133.
27 Dewis P, Newman M, Ratcliffe WA, et al: Does testosterone affect the normal menstrual cycle? Clin Endocrinol 1986;24: 515-521.
28 Mildenberger P, Lotz R, Bierhoff E, et al: Tumoreinbruch in die Vena cava inferior bei Nebennierenkarzinomen. Fortschr Roentgenstr 1987;147:160-165.
29 Klöti R, Müller J, Floersheim Y, et al: Androgen produzierender Hilsuzelltumor (Leydig-Zell-Tumor) des Ovars: Kasuistischer Beitrag zur Differentialdiagnose der Virilisierung der erwach- senen Frau. Schweiz Med Wochenschr 1980;110:453-458.
30 Gross MD, Wilton GP, Shapirto B, et al: Functional and scinti- graphic evaluation of the silent adrenal mass. J Nucl Med 1987; 28:1401-1407.
31 Arnold A, Stounton CE, Kim HG, et al: Monoclonality and abnormal parathyroid hormone genes in parathyroid adenomas. N Engl J Med 1988;318:658-662.
32 Burr IM, Sullivan J, Graham T, et al: A testosterone secreting tumour of the adrenal producing virilization in a female infant. Lancet 1973;ii:643-644.
33 Kable WT, Yussman MA: Testosterone-secreting adrenal adeno- ma. Fertil Steril 1979;32:610-611.
34 Mack E, Sarto GE, Crummy AB, et al: Virilizing adrenal gan- glioneuroma. JAMA 1978;239:2273-2274.
35 Dolinar R, Burch WM: Testosterone producing adrenal ade- noma in woman with normal urinary 17-ketosteroid levels. JAMA 1983;250:2504-2505.
36 Spaulding SW, Masuda, T, Osawa Y: Increased 17a-hydroxyste- roiddehydrogenase activity in a masculinizing adrenal adenoma in a patient with isolated testosterone overproduction. J Clin Endocrinol Metab 1980;50:537-540.
37 Kelly TR, Mayors DJ, Boutsicaris PS: Adrenal adenoma: Iso- lated testosterone secretion. Am Surg 1982;48:604-606.
38 Larson BA, Vanderlaan WP, Judd HL, et al: A testosterone pro- ducing adrenal cortical adenoma in an elderly woman. J Clin Endocrinol Metab 1976;42:882-887.
Received: May 18, 1988 Accepted after revision: September 16, 1988
W. Sorgo, MD Department of Pediatrics University of Ulm Prittwitzstr. 43 D-7900 Ulm (FRG)
Downloaded by: