A 6-year-old girl (K.G.) was referred to the Pediatric Endocrinology Clinic at Oregon Health Sciences University (OHSU) for evaluation of rapidly advancing puberty. Parents first noted the development of acne six months prior to her initial visit. This was followed by the onset of breast development three months prior to the visit. Finally, 1 month prior to her initial evaluation, K.G. developed pubic hair. Menstrual bleeding began and continued for the two weeks prior to her endocrinologic evaluation. Further history revealed that K.G. had experienced mood swings recently and she had been troubled by unusually oily hair and skin. Her energy level and appetite had remained normal. She denied any history of headaches or visual complaints. No history of polyuria, polydipsia or nocturia was noted. As well, she denied any abdominal discomfort, nausea or vomiting and change in her bowel habits.
Past history was unremarkable. Her growth & development including early milestones, vision & hearing were all normal. She was in Kindergarten and did well academically. Of note is that she was the tallest child in her class. She was on no medications.
Family history was significant for Mother age 32 years with a history of Type I Neurofibromatosis (NF1).The mother’s menses occurred at 12 years. Father, age 33 years, had a history of kidney stones. Midparental height was 5 ft. 6 1/2 inches (75–90%). There was no history of consanguinity, adrenal or thyroid disease. A paternal grandmother had Type 2 Diabetes.
On physical exam, K.G. had a somewhat deepened voice, appeared tall for her age and had marked facial acne. No features consistent with Cushing’s syndrome including moon facies, plethora, striae or unusual downy hair on the face or forehead were noted. Growth parameters showed weight 26.5 kg (95%) and height 128.1 cm (>95% ). Growth velocity was unavailable at initial visit. The blood pressure was 121/67. Fundoscopic examination was normal, no thyroid was palpable and no significant lymphadenopathy was present. Respiratory, cardiovascular and abdominal exams were all normal. No abdominal mass was present. She had Tanner III Breast and Tanner III Pubic hair development. Mild clitoromegaly was noted. No posterior labial fusion was present. Four cafe-au-lait spots were noted on her left leg. Neurological exam was entirely normal.
Bone Age (Greulich and Pyle Method) 71/2 years (+11/2 SD) at chronological age 6 years.
Pelvic ultrasound demonstrated normal prepubertal ovaries and a postpubertal uterus. Abdominal ultrasound showed a rounded left-sided adrenal mass measuring 3.7 × 4.1 × 3.7 cm without calcifications. No liver metastases, celiac or SMA lymphadenopathy, ascites or splenomegaly was seen (see Fig. 1).
Contrast CT Scan showed normal liver, spleen, gallbladder, pancreas and kidneys. A normal right adrenal gland was noted. A well-circumscribed left adrenal mass was noted. The size of the mass caused some posterior displacement of the left kidney and anterior displacement of the body and tail of the pancreas. No retroperitoneal adenopathy was identified (see Fig. 2).
Initial laboratory testing was done including measurements of electrolytes, 24 h urine 17-ketosteroids, serum cortisol, 24h urine free cortisol, serum estrone and serum estradiol. Initial and follow-up laboratory tests are summarized in Table 1.
At operation, by exploratory laparotomy, a left-sided adrenal adenoma that was encapsulated without any evidence of invasion of surrounding structures was found. This tumor was completely removed. Stress doses of hydrocortisone were administered post-operatively for 4 days. These were discontinued when the preoperative 24 hour urine free cortisol was confirmed to be normal.
Pathology on gross examination revealed tissue weighing 22 grams with dimensions of 5.3 × 3.5 × 2 cm. Sectioning revealed distinct residual adrenal gland measuring 3.3 × 1.7 × 0.3 cm. This residual adrenal tissue consisted of a yellow appearing cortex and a central gray medulla. There was an attached mass arising from the cortical aspect of this specimen which contained a large amount of friable, rubbery, granular, tan-red material.
At follow-up visits up to 15 months post-operative, K.G. remains well. She has had complete regression of both breast and pubic hair development, no complaints of further acne or vaginal bleeding and growth velocity has slowed to 4.6 cm/year (normal for age). In addition, her follow-up laboratory values remain in the normal prepubertal range (see Table 1.)
Precocious puberty is typically gonadotropin-dependent and caused by a disturbance of the hypothalamicpituitary axis. Idiopathic true precocity accounts for 95% of the true precocity in girls and is due to premature triggering of the normal pubertal mechanism. Of the remaining cases of early central puberty, any type of intracranial abnormality may be responsible; these include congenital causes such as cerebral palsy or hydrocephalus and acquired causes such as irradiation, trauma, chronic inflammatory disorders or masses in the hypothalamic area. No clinical symptoms to suggest any of these processes were present in the above patient.
Of note, the precocious puberty associated with neurofibromatosis type I, an autosomal dominant disorder, usually results from an optic glioma or hamartoma. Also, pineal tumors may cause premature puberty by an absence of a normal pineal inhibitory factor which holds puberty in check rather than a destructive effect on the inhibitory tracts themselves. Finally, a peripheral endocrine disorder that results in advancement of the bone age to 10 to 12 years may cause precocious puberty by advancement of somatic maturation. This degree of advancement in bone age was missing in our patient.
Gonadotropin-independent precocious puberty is much less common in girls. Consideration however, was given to whether our patient had McCune-Albright syndrome, a syndrome of premature puberty, irregularly shaped café au lait spots and polyostotic fibrous dysplasia. Precocious puberty in this disorder is due to autonomous functioning ovarian luteinized follicular cysts. The underlying defect in this disorder is a somatic activating mutation in the α subunit of a G protein. Patients with this condition often have menstrual bleeding early in the course similar to our patient, but do not usually have early pubic hair development. Other ovarian causes of precocious puberty were also considered such as a granulosa cell tumor which is the second most common ovarian feminizing neoplasm in girls, a feminizing ovarian cyst or masculinizing tumors like Leydig-Sertoli cell tumors or arrhenoblastomas. Again this did not fit the case presentation of both signs of feminization and virilization occurring together.
Finally, adrenal disorders as a cause of precocious puberty were considered. Our case did not fit the presentation of congenital adrenal hyperplasia which results in signs of virilization without breast development or early menses. Adrenocortical tumors usually cause a rapid virilizing disorder due to high androgens with or without symptoms related to cortisol excess. However, rarely adrenal tumors can cause feminization and when feminization and virilization coexist due to such a tumor, the clinical presentation is identical to that of true precocious puberty. Our case is one of these very rare examples of how an adrenal tumor may present resembling true precocious puberty.
Adrenocortical tumors are rare in childhood. They are classified as functional or nonfunctional. In contrast to adults, most of the adenomas show hormonal secretion in children. Of those adrenocortical tumors which present in childhood, 41.3% occur before 2 years of age and 74.1% appear before age of 5 years. In a retrospective study of 54 children with adrenal tumors , age of diagnosis ranged from 5 months to 16 years (median, 4 years) with two peaks of frequency in the first years of life and at 12–14 years of age . Studies done by Sandrini et al. of 58 Brazilian children, Lack et al. of 30 patients and Michalkiewicz et al. of 20 of 67 children who had small adrenocortical tumors (weighing < 100g) also reported a similar mean age at diagnosis of 4.3 yr, 5 years and 2 years respectively [4–6].
Nader et al. reports the frequency of adrenocortical tumors as one case in every 1.7 million people which accounts for 0.6% of all childhood tumors . In addition, Hartley et al., based on data from the Manchester Children’s Tumor Registry and others have found that adrenocortical carcinoma specifically comprised 0.3% of all childhood malignant tumors [8,9]. In stark contrast to this, the annual incidence of adrenocortical tumors is much greater in Southern Brazil with an annual incidence of 3.4–4.2/million children below the age of 15 years. It is thought that environmental pollutants may be responsible for the increased incidence of adrenocortical tumors found in Southern Brazil . Alternatively, these patients may harbor a genetic defect based on findings of a gain of chromosomal region 9q34 which was found in 8 out of 9 tumors from Brazilian children. Thus, this chromosomal region may contain an oncogene important for adrenocortical tumor formation .
Hayes et al. has reported a higher incidence of adrenal tumor formation in females (female:male ratio = 2.3:1) as has Humphrey et al. finding a 2.1:1 female:male ratio . However, other investigators have reported equal incidence in boys and girls . A consistent finding in all studies, however, is that the majority of patients with adrenocortical tumors are Caucasian.
Clinical epidemiologic studies have shown a number of interesting associations in pediatric adrenocortical neoplasms. Of note, adrenocortical tumors have been found to be associated with certain congenital abnormalities including: Hemihypertrophy, Beckwith-Wiedemann syndrome, urinary tract abnormalities, the Li-Fraumeni Syndrome in which these patients have a missing p53 tumor suppressor gene, astrocytoma, cutaneous lesions, and abnormalities of the contralateral adrenal gland.
It is interesting to speculate on an additional possible association between NF1 and the formation of adrenocortical neoplasms. A unique feature of this case is the history of NF1 in this patient’s mother and the presence of 4 café-au-lait spots on physical examination of this patient. Sartori et al. reported the occurrence of an adrenocortical adenoma causing Cushing’s syndrome in a 13-year-old girl with neurofibromatosis (NF) and reported three other previous cases of adrenocortical tumors in patients with NF . This suggested an association between NF and adrenocortical tumors. Similar to the situation in Retinoblastoma, the NF1 gene is thought to be a tumor suppressor gene and loss of both copies of this gene then would predispose to malignant transformation. Since neurofibromin is expressed in the adrenal gland, Gutmann et al. examined the expression of neurofibromin in five pheochromocytomas as well as two additional adrenocortical tumors from NF 1 patients . In all tumors examined, no neurofibromin was detected as shown by loss of heterozygosity analysis and immunoprecipitation and Western Blot methodology. Therefore, this suggests that the NF1 gene product may be involved in the pathogenesis of adrenal and other tumors.
Often the generally healthy appearance, increased growth pattern and lack of an abdominal mass deters pediatricians from thinking about the possibility of a malignancy in these patients. A delay from initial symptoms to diagnosis was median 0.5 years (0.1–5.5 years) in one series . Other studies have reported similar intervals between initial symptoms and diagnosis of median of 5.5 months, 10 months, 9 months, and 8 months [references 4,5,6,15, respectively].
The initial presenting symptoms of many series of children with adrenocortical tumors has been reviewed by various authors. Most children present with symptoms and signs of androgen excess, hypercortisolism or rarely, hyperaldosteronism or feminization [16,17]. Interestingly, children have a higher incidence of virilization (72%) as opposed to adults whose predominant symptom is Cushing’s syndrome (60%) . The most notable features of virilization in female children are increased muscle mass, facial hair, deepening of the voice, the appearance of pubic hair and clitoral enlargement. In male children, virilizing symptoms include isosexual precocity with penile enlargement and pubic hair. The most frequent manifestations of hypercortisolism in children are hypertension, truncal obesity, moon facies and plethora.
Sandrini et al. reviewed 58 Brazilian children with adrenocortical tumors and found that virilization alone was the presenting feature in 40% of patients. Isolated signs of Cushing’s syndrome were rare (3% of patients) but often accompanied signs of virilization in 50% of cases. An elevated blood pressure was also seen in 55% of these patients. Also, a palpable abdominal mass was present in 48% of these patients with equal distribution between the left and right adrenal glands . Teinturier also looked at a series of 54 patients presenting with adrenal tumors. Thirty three children (61%) were referred for endocrine symptoms and 21 (39%) were referred because of an abdominal mass . Upon careful examination of these same children, 44 children (81%) actually did have presenting endocrine manifestations including: pubic hair in prepubertal children or hirsutism and acne in teenagers in 41 patients (76%) with or without enlarged penis (30% of boys) or clitoromegaly (21% of girls), Cushing syndrome in 8 patients (15%) and gynecomastia in three boys. An abdominal mass was present in 31 patients (57%)  (see Table 2). In addition to these more usual presentations, Conn syndrome with hypokalemia and hypertension may signify a rare aldosterone-secreting tumor . Finally, feminization as a initial presenting symptom similar to that seen in our case presentation is very rare but isolated cases have been reported [20–24]. One other case report described an adrenocortical adenoma with hypersecretion of estradiol, androgens and cortisol in a 6-year-old boy . Only two other cases of an adrenocortical adenoma presenting as isosexual female precocity was found upon extensive review of the literature [26,27]. One case of a feminizing adrenal adenoma was shown to contain increased levels of aromatase activity as measured by RT-PCR in the tumor sample .
Diagnosis may be confirmed by measurement of timed urinary collections for 17-Ketosteroids, 17-OHCS, DHEA and free cortisol as well as serum measurements of DHEA, DHEAS, cortisol, testosterone and estrogen. An elevated urinary 17-ketosteroid result often provides a pivotal clue in the diagnosis of adrenocortical tumors and was found to be abnormal in >99% of patients reported . The next most useful marker is serum DHEA sulfate levels as these were abnormal in 90% of 20 out of 58 patients reported by Sandrini et al. . Of note is that all cases with clinical signs of excessive glucocorticoids reported by Sandrini were found to have elevated urinary 17-hydroxycorticosteroid levels and thus, this is a sensitive means to diagnosis Cushing’s syndrome in these patients . Furthermore, no tumors which produce excess cortisol should be suppressible with the administration of exogenous glucocorticoids .
Radiological studies such as abdominal sonography or CT scanning or MRI are useful in confirmation of the diagnosis and in accurate localization of the adrenal tumor. Of note is that in 28 patients reported by Sandrini et al. who were evaluated by both ultrasound and CT scan, the sonogram missed an adrenal tumor in 3 cases (11%) . Therefore, all patients suspected of having an adrenal tumor should be evaluated by CT or MRI and very rarely, when small tumors are present and there is a high index of clinical suspicion, exploratory laparotomy may be indicated.
The histologic classification of adrenocortical tumors, is very challenging . Childhood tumors are clearly different from those occurring in adults. Unlike adrenal tumors in adult patients where nuclear grade, mitotic rate, atypical mitoses, percent clear cells, architecture, microscopic necrosis and invasion of venous, sinusoidal and capsular structure indicate malignancy, only size was found to be predictive of malignancy in pediatric tumors. All pediatric tumors weighing more than 500 g were malignant while all but one weighing less than 500 g were benign. Furthermore, the remaining features were present in both benign and malignant tumors and thus, not useful distinguishing factors . Bugg et al. also examined the prognostic value of histologic classification, ploidy, proliferative index and size (tumor weight or greatest diameter) in a series of pediatric adrenocortical neoplasms from 54 Brazilian children. This study concluded that histologic type and tumor weight were in fact the most reliable predictors of outcome in pediatric adrenocortical neoplasms .
Complete surgical excision is the only mode of therapy which provides a chance of prolonged disease-free survival. Overall objectives in the treatment of these tumors include surgical removal of the tumor and returning steroid secretion back to prepubertal values. Cortisol replacement is almost always needed perioperatively because the contralateral adrenal gland is frequently atrophic and suppressed by tumor cortisol production. Careful attention to electrolyte balance, hypertension, surgical wound care and infectious complications is imperative in all cases. The role of chemotherapy in the management of childhood adrenocortical tumors is still uncertain.
Prognosis of these adrenocortical tumors is unclear to date. In the larger series of pediatric adrenocortical tumors, mortality rates are reported between 50 and 57% [3,5,25,31,32]. Of the 54 patients reported by Sandrini et al., 24 (44%) died and 30 (56%) were disease-free for a median of 63 months from diagnosis. In the unfortunate circumstance that tumors do recur, they tend to do so within 1 to 2 years of initial resection. In search of predictors for outcome, the presence of metastases at diagnosis or failure to completely resect the tumor is associated with extremely dismal outcomes. Sandrini et al. have proposed a staging classification of adrenocortical neoplasms as shown (Table 3). This is based on analysis of variables in patients without metastatic disease which were associated with worse outcome: age greater than 3.5 years, greater than 6 month delay between initial symptoms and diagnosis, greater than 4 mg/m2/day urinary excretion of 17-OHCS, tumor volume greater than or equal to 200 cm3 and tumor weight greater than 80 g . A similar independent analysis done by another group found these same factors as predictive of unfavorable outcome .
Finally, monitoring of clinical signs and adrenal steroid hormone levels are the most accurate means of detecting recurrence. Survivors of adrenocortical tumors need to be followed long-term to ensure appropriate growth and development. Surprisingly, although most patients do have an element of advanced physical and skeletal maturity and are therefore are at risk for adult short stature and early puberty, these patients usually have a “catch-down” effect more pronounced for bone age than for growth velocity so that their final predicted adult height does not differ from their genetic potential [4,30].
This case illustrates a novel case of an adrenal tumor capable of producing both estrogens and androgens. It highlights important clinical features such as a rapid clinical course, mild clitoromegaly and severe acne which prompted an aggressive pursuit of the underlying etiology. In light of recent recommendations regarding redefining the clinical definition of precocious puberty, it is still imperative to evaluate every case of precocious puberty on an individual basis and search for any worrisome features as were found in this case.
1. Herman-Giddens ME, Slora EJ, Wasserman RC, et al.: Secondary sexual characteristics and menses in young girls seen in office practice: a study from the Pediatric Research in Office Settings network. Pediatrics 1977; 99:505–12.
2. Kaplowitz PB, Oberfield SE, and the Drug and Therapeutics and Executive Committees of the Lawson Wilkins Pediatric Endocrine Society: Reexamination of the age limit for defining when puberty is precocious in girls in the United States: implications for evaluation and treatment. Pediatrics 1999; 104:936–41.
3. Teinturier C, Brugieres L, Lemerle J, et al.: Adrenocortical neoplasms in childhood: a retrospective study of 54 cases. Arch Pediatr 1996; 3:235–40.
4. Sandrini R, Ribeiro RC, DeLacerda L: Extensive personal experience: childhood adrenocortical tumors. J Clin Endocrinol Metab 1997; 82:2027–31.
5. Lack EE, Mulvihill JJ, Travis WD, et al.: Adrenal cortical neoplasms in the pediatric and adolescent age group. Clinicopathologic study of 30 cases with emphasis on epidemiological and prognostic factors. Pathol Ann 1992; 27:1–53.
6. Michalkiewicz EL, Sandrini R, Bugg MF, et al.: Clinical characteristics of small functioning adrenocortical tumors in children. Med Pediatr Oncol 1997; 28:175–8.
7. Nader S, Hickey RC, Sellin RV, Samaan NA: Adrenal cortical carcinoma. A study of 77 cases. Cancer 1983; 52:707–11.
8. Hartley AL, Birch JM, Marsden HB, et al.: Adrenal cortical tumors: epidemiological and familial aspects. Arch Dis Child 1987; 62:683–890.
9. Daneman A, Chan HSL, Martin J: Adrenal carcinoma and adenoma in children: a review of 17 patients. Pediatr Radiol 1983; 13:14–18.
10. Figuiredo BC, Stratakis CA, Sandrini R, et al.: Comparative genomic hybridization analysis of adrenocortical tumors of childhood. J Clin Endocrinol Metab 1999: 84:1116–21.
11. Humphrey GB, Pysher T, Holocombe J, et al.: Overview of the management of adrenocortical carcinoma. In Adrenal and Endocrine Tumors in Children, edited by Humphrey GB, Grindey GB, Dehner LP, et al., pp. 349–57, Boston: Martinus Nijhoff Publishers 1983.
12. Teinturier C, Pauchard MS, Brugières L, et al.: Clinical and prognostic aspects of adrenocortical neoplasms in childhood. Med Pediatr Oncol 1999; 32:106–11.
13. Sartori P, Symons JC, Taylor NE, Grant DB: Adrenal cortical adenoma in a 13 year old girl with neurofibromatosis. Acta Paediatr Scand 1989; 78:476–8.
14. Gutmann DH, Cole JL, Stone WJ, Ponder BAJ, Collins FS: Loss of neurofibromin in adrenal gland tumors from patients with neurofibromatosis type I. Genes Chromosom Cancer 1994; 10:55–8.
15. Mayer SK, Oligny LL, et al.: Childhood adrenocortical tumors: case series and reevaluation of prognosis—a 24-year experience. J Pediatr Surg 1997; 32:911–5.
16. Lee PDK, Winter RJ, Green OC: Virilizing adrenocortical tumors in childhood: eight cases and a review of the literature. Pediatrics 1985; 76:437–44.
17. Morales L, Rovira J, Rottermann M, et al.: Adrenocortical tumors in childhood: a report of four cases. J Pediatr Surg 1989: 276–81.
18. Mendonca BB, Lucon AM, Menezes CAV, et al.: Clinical, hormonal and pathological findings in a comparative study of adrenocortical neoplasms in childhood and adulthood. J Urol 1995; 154:2004–9.
19. Schmitt K, Frisch H, Neuhold N, et al.: Aldosterone and testosterone producing adrenal adenoma in childhood. J Endocrinol Invest 1995; 18:69–73.
20. Bacon GE, Lowrey GH: Feminizing adrenal tumor in a six-year-old boy. J Clin Endocrinol Metab 1965; 25:1403–6.
21. Bhettay E, Bonnici F: Pure estrogen-secreting feminizing adrenocortical adenoma. Arch Dis Child 1977; 52:241–3.
22. Leditschke J, Arden F: Feminizing adrenal adenoma in a 7 year old male. Aust Paediatr J 1974; 10:217–21.
23. Mosier DH, Goodwin WE: Feminizing adrenal adenoma in a 7 year old boy. Pediatrics 1961; 27:1016–21.
24. Smith AH: A case of feminizing adrenal tumor in a girl. J Clin Endocrinol Metab 1958; 18:318–22.
25. Balakumar T, Perry LA, Savage MO: Adrenocortical adenoma—an unusual presentation with hypersecretion of oestradiol, androgens and cortisol. J Pediatr Endocirinol Metab 1997; 10:227–9.
26. Scully RF, Galdabini JJ, McNeely BU: Case records of the Massachusetts General Hospital: weekly clinicopathological exercises. New Engl J Med 1979: 300:1322–8.
27. Phornphutkul C, Harel Z, Goodwin G, Tracy T, et al.: Aromatase expression in a feminizing adrenal adenoma. Pediatr Research 2000; 47:137A.
28. Street ME, Weber A, Cama Cho-Hübner C, et al.: Girls with virilisation in childhood: a diagnostic protocol fro investigation. Clin Pathol 1997; 50:379–83.
29. Cagle PT, Hough AJ, Pysher H, et al.: Comparison of adrenal cortical tumors in children and adults. Cancer 1986; 57:2235–7.
30. Bugg MJ, Ribeiro RC, Roberson PK, et al.: Correlation of pathologic features with clinical outcome in pediatric adrenocortical neoplasia. Am J Clin Pathol 1993; 101:625–9.
31. Sabbaga CC, Avilla SG, Schulz C, et al.: Adrenocortical carcinoma in children: clinical aspects and prognosis. J Pediatr Surg 1993; 28:841–3.
32. Lefevre M, Gerard-Marchant R, Gubler JP, et al.: Adrenal cortical carcinoma in children: 42 patients treated from 1958 to 1980 at Villejuif. In Adrenal and Endocrine Tumors in Children, edited by Humphrey GB, Grindey GB, Dehner LP, et al., pp. 265–76, Boston, Martinus Nijhoff Publishers 1983.
33. Ribeiro RC, Neto RS, Schell MJ, et al.: Adrenocortical carcinoma in children: a study of 40 cases. J Clin Oncol 1990: 67–74.