Review

TRENDS in Endocrinology and Metabolism

Vol.18 No.4

Paediatric Cushings syndrome: epidemiology, investigation and therapeutic advances

Helen L. Storr, Li F. Chan, Ashley B. Grossman and Martin O. Savage
Department of Endocrinology, William Harvey Research Institute, Barts and the London, Queen Marys School of Medicine and Dentistry, John Vane Science Centre, Charterhouse Square, London, EC1M 6BQ, UK

Cushings syndrome (CS), which is caused by excessive circulating glucocorticoid concentrations, is rare in the paediatric age range but presents a diagnostic and therapeutic challenge. Most paediatric endocrinologists have limited experience of managing children or adolescents with CS and thus benet from close consultation with colleagues who treat adult patients. A protocol for investigation is required that broadly follows the model for adult patients. Here, the epidemiology and diagnosis of different causes of CS are discussed according to typical age of presentation. Treatment strategies for adrenocorticotrophic hormone (ACTH)-independent and ACTHdependent CS are described and critically appraised. The management of paediatric CS patients after cure also presents challenges for optimizing growth, bone health, reproduction and body composition from childhood into and during adult life. Introduction Cushings syndrome (CS) is very rare in childhood and adolescence, compared with the more common endocrine pathologies such as disorders of growth, puberty and thyroid which combine to make up the major part of paediatric endocrine practice. CS is a clinical syndrome caused by excessive circulating glucocorticoid concentrations resulting from either endogenous secretion or exogenous administration and can present difcult diagnostic and therapeutic challenges for the clinician [1,2]. Here, we focus on areas that particularly characterize paediatric CS. We discuss epidemiology, investigation and advances in treatment that have relevance to the clinical management of patients. We also emphasize that very few, if any, paediatric endocrinology units have sufcient experience to manage CS in isolation, and that consultation and joint decision-making with more experienced adult endocrinology units will be benecial to the care of the patient. Classication of paediatric Cushings syndrome CS in childhood and adolescence can be classied into two groups of adrenocorticotrophic hormone (ACTH)-independent and ACTH-dependent causes (Box 1). Although CS caused by unphysiological glucocorticoid replacement
Corresponding author: Savage, M.O. (m.o.savage@qmul.ac.uk). Available online 6 April 2007.
www.sciencedirect.com

comprises the commonest paediatric cause, such children are rarely referred for endocrine management and remain largely under the care of the paediatrician treating their primary condition. However, it is always important to enquire about possible exogenous sources such as inhaled or topical corticosteroids or the use of unusual food additives and supplements. Epidemiology CS can present throughout childhood and adolescence, but certain aetiologies occur more frequently at particular ages. A summary of aetiologies related to age is shown in Figure 1. The different diagnostic categories from the neonatal period to adolescence are discussed below. McCune-Albright syndrome CS in infancy is usually associated with McCune-Albright syndrome (MAS), a sporadic heterogeneous disorder caused by activating mutations of Arg201 in the guanine-nucleotide-binding protein (G protein) a subunit [3]. In infancy, MAS, occurring predominantly in females, can present with CS, often co-existing with additional endocrine abnormalities such as hyperthyroidism or precocious puberty [4,5]. Non-endocrine features, such as cutaneous stigmata, nephrocalcinosis, hepatobiliary and cardiac defects can accompany CS. The course of CS can be variable but is usually severe and potentially life threatening, requiring bilateral adrenalectomy. Occasionally it can resolve spontaneously [5]. The histological appearance of the adrenal tissue shows nodular adrenocortical hyperplasia [5,6]. Adrenocortical tumours Adrenocortical tumours (ACTs) comprise only 0.30.4% of neoplasms in children, but are an important cause of paediatric CS [7]. Adrenocortical carcinoma (ACC) should be suspected in any child with CS and an adrenal mass [8]. Much has been learnt from the large ACT series from Southern Brazil, where the incidence of paediatric ACC is 3.44.2 per million children. [9,10]. ACC can also be associated with the Li-Fraumeni syndrome, and in a study of 36 ACC patients and their families, 35 patients had an identical germline point mutation of the p53 tumour suppressor gene (TP53) encoding an R337H amino acid substitution [9]: the presence of this mutation did not inuence

1043-2760/$ see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.tem.2007.03.005

168

Review

TRENDS in Endocrinology and Metabolism Vol.18 No.4

Box 1. Classification of paediatric Cushings syndrome

ACTH-independent 1. Exogenous glucocorticoid administration (tablets, nose drops, nasal spray, skin cream) 2. Adrenocortical tumour (adenoma or carcinoma) 3. Primary adrenocortical hyperplasia  Primary pigmented nodular adrenocortical disease (PPNAD)  Macronodular adrenal hyperplasia (AIMAH)  McCune-Albright syndrome ACTH-dependent 1. Cushings disease (ACTH-secreting pituitary adenoma) 2. Ectopic ACTH syndrome

[17] and a carcinoid tumour arising in the duodenum [18] have also been reported. Out of eight carcinoid tumours causing paediatric EAS, six have been in females and the median age for all cases was 9.5 years. A clear cell sarcoma has also been reported to cause EAS [19], as has a malignant neuroendocrine tumour of the pancreas in a 16-year-old female resulting from secretion of ACTH precursors [20]. A further rare source of EAS is Wilms tumour, in younger children [21], and adrenal neuroblastoma which has been reported in several infants [22]. Primary nodular adrenal hyperplasia Primary bilateral adrenocortical diseases are rare, two are implicated in the pathogenesis of paediatric ACTH-independent CS. The rst is primary pigmented nodular adrenocortical disease (PPNAD) or micronodular adrenal disease, which is a congenital disorder: most cases (95%) are associated with the multiple endocrine neoplasia (MEN) syndrome known as Carney complex. The second disease is ACTH-independent macronodular adrenal hyperplasia (AIMAH) or massive macronodular adrenocortical disease (MMAD), which is rarely reported in the paediatric population. Other forms of bilateral adrenocortical hyperplasia include the adrenal lesions in McCuneAlbright and MEN1 syndromes. Primary pigmented adrenocortical disease Primary pigmented adrenocortical disease (PPNAD) typically occurs in late adolescence or early adulthood [2325]. The adrenal pathology consists of multiple, small, pigmented, adrenocortical nodules surrounded by internodular cortical atrophy [26,27]. The cortical atrophy results from autonomous function of the adrenal nodules and the resulting suppression of ACTH secretion. This histological appearance is pathognomonic for PPNAD, although adrenal pigmentation can be absent [28,29]. The hypercortisolaemia of PPNAD can rarely be subclinical [23] or associated with several different clinical phenotypes, and paediatric patients might present

prognosis in this group [11]. The inuence of TP53 mutations on adrenocortical tumourigenesis remains unclear and might depend on complex genetic or environmental interactions [12]. ACT occurs most commonly in children under four years of age and is usually associated with virilization (56% in the Brazilian series). Mixed hormone secretion, including cortisol, was the next commonest cause (29.2%); pure cortisol-secreting ACT was rare, making up 5.5% and 3% of patients in two series [10,13]. Isolated CS tended to occur in older children (median age 12.6 years) with an equal sex incidence. Hypertension was commonly present. Older patients with CS or mixed hormone secreting tumours had lower survival rates than younger children [10]. Ectopic ACTH syndrome Ectopic ACTH syndrome (EAS) is extremely rare in the paediatric age range, occurring much less frequently than its approximate 15% prevalence in adult ACTH-dependent CS [14]. However, paediatric EAS has been well documented and some paediatric cases are included in the large published series [1416]. Carcinoid tumours predominate as causes of paediatric EAS in children. Most tumours are bronchial or thymic, but oncocytic renal carcinoid

Figure 1. Different aetiologies of paediatric Cushings syndrome from the literature (n = 398 cases) shown at ages of peak incidence (boxes).
www.sciencedirect.com

Review

TRENDS in Endocrinology and Metabolism

Vol.18 No.4

169

differently from adults [30,31]. The classical features of CS can be absent in childhood [32], but in our series, all patients displayed the typical features of CS, including hypertension and virilization [28]. Another variant known as cyclical or periodic Cushing syndrome is described [29,33] in which normal cortisol production is interrupted by periods of hypercortisolaemia. PPNAD is frequently associated with the Carney complex (CNC; MIM 160980) [27], an autosomal dominant MEN syndrome [34,35] characterized by pigmented spots, cardiac myxomas, and endocrine and non-endocrine tumours [25,27]. PPNAD is the most frequent presentation of CNC in children and young adults [23,24]. Cushings disease Pituitary-dependent Cushings disease (CD), dened as hypercortisolaemia caused by an ACTH-secreting corticotroph adenoma, is the commonest form of CS in childhood and adolescence, accounting for 7580% of cases [1,36,37]. The pituitary tumour is almost always a microadenoma [38]; only one macroadenoma occurred out of 32 paediatric cases in the series from our centre [39]. However, this has been reported very rarely [40] and the tumour might even invade the cavernous sinus [41]. Pituitary macroadenoma might also be an early manifestation of MEN1 [42]. The commonest age of presentation of paediatric CD is during the adolescent or preadolescent years. The median age of presentation of 141 cases taken from the literature was 14.1 years (Figure 1), and the youngest child in our own series was aged 6.2 years. In adults, CD has a female preponderance [43]. Until recently, no comment had been made about sex distribution of CD in children [44,45]. Sex was analysed at diagnosis in 50 CD patients aged from 6 to 30 years and a signicant predominance of males was found in the prepubertal patients [46]. There were similar incidences in both sexes during puberty and an increasing predominance of females in the post-pubertal patients. This report was the rst to describe this male predominance in young CD patients, although examination of cases in the large series from the NIH [36] reveals the same phenomenon. There is no clear explanation for this, although the oestrogenic milieu during puberty in females might be related to the relative increase in females with CD during and following adolescence. Clinical presentation As CD is the most frequent cause of CS after the preschool years, a brief description of common clinical features is warranted. The recognition of features alerting the clinician to the diagnosis of CD is of crucial importance in effective diagnosis and treatment. Most children and adolescents with CD have a typical Cushingoid appearance. A subtle or subclinical presentation or even cyclical features are uncommon. However, parents and general practitioners frequently fail to recognize the pathological nature of the change of the childs appearance. The mean duration of symptoms before diagnosis in our 32 CD patients was 2.5 years (range 0.56.6 years). The facial appearance was always changed and all of our patients experienced weight gain. Striae were present in 53% of the
www.sciencedirect.com

patients, and were more frequent in the older patients. It was not unusual for a young child with CD to present with obesity and poor growth, without the classical features of plethora, hirsutism, acne and striae. Additional features were hypertension (47%), emotional lability (53%) and fatigue (59%). Muscle weakness and easy bruising were rare. Characteristics of growth Short stature (height less than 2.0 SD) was present in 40% of patients and growth velocity was subnormal. One of the most striking features was the contrast between height SDS values, which were almost always below the mean, and body mass index (BMI) SDS values which were consistently above it (Figure 2), as previously reported [1]. Height and BMI SDS values were recently compared in 29 patients with CD and 44 age-matched subjects with simple obesity. There was a signicant difference in the ratio of these two variables between the two groups [47]: height was increased in simple obesity and decreased in CD. Bone age (BA) at diagnosis was delayed in 15 out of 17 patients with CD (mean delay 2.0 years; range 0.5 to 4.1 years) and correlated negatively with height SDS (r = 0.70; P < 0.01), duration of symptoms (r = 0.48; P = 0.05) and age at diagnosis (r = 0.48; P = 0.05) [48]. Puberty development There are few reports of pubertal development in CD, although virilization with pseudo-precocious puberty is recognized [1,36]. Puberty was analysed in 27 CD patients, of whom 12 had abnormal virilization, dened as unusual advance of Tanner pubic hair stage compared with

Figure 2. Height (blue) and body mass index (BMI; red) SDS values in 32 paediatric patients with Cushings disease. The dotted line indicates the SDS value below which patients are significantly shorter than average. These data are from our unit at the William Harvey Research Institute.

170

Review

TRENDS in Endocrinology and Metabolism Vol.18 No.4

testicular volume or breast development [49]. In the virilized patients, serum androstenedione, dehydroepiandrosterone (DHEAS) (as previously reported [50]) and testosterone SDS were higher (P = 0.03, 0.008, 0.03, respectively) than in subjects without abnormal virilization, and sex-hormone binding globulin (SHBG) SDS values were lower (P = 0.006). Gonadotrophin levels were subnormal in the patients who had commenced true puberty, indicating a suppressive effect of chronic hypercortisolaemia. Investigation of Cushings syndrome The investigation of patients with suspected CS in adults [51,52] and children [1] has been extensively reviewed. We therefore highlight some aspects that we have found helpful during the management of 47 paediatric CS patients over the past 25 years. The scheme of investigations in children should be based on the adult protocol [51]. The protocol consists initially of conrmation or exclusion of the diagnosis of CS followed by investigations to dene the aetiology (Box 2). Conrmation or exclusion of Cushings syndrome As initial investigations, three consecutive 24 h urine collections are performed for urinary free cortisol (UFC) followed by measurement of serum cortisol at three time-points (09.00h, 18.00h and midnight [sleeping]) to assess circadian rhythmicity. The midnight cortisol in the sleeping child should be <50 nmol l1, although some children can reach their cortisol nadir earlier than midnight. A low-dose dexamethasone suppression test (LDDST) is then performed, using the adult dose regimen of 0.5 mg 6 hourly (at 09.00, 15.00, 21.00 and 03.00 h) for 48 h, unless the child weighs <40 kg which is unusual when the NIH-recommended dose of 30 mg kg1 day1 is used [36]. In the LDDST, blood is taken for serum cortisol basally and then at 0, 24 and at 48 h, when it is normally undetectable (<50 nmol l1). These tests performed individually, and particularly in combination, have a high sensitivity for CS and an even higher specicity for the exclusion of this diagnosis. It might be that in the future the use of midnight salivary cortisol will be of value as a screening test, but normative values for children are not currently available.

Denition of the aetiology of Cushings syndrome When the presence of CS has been conrmed, the priority is to establish whether the patient has ACTH-dependent or ACTH-independent CS. This is most easily determined by measurement of basal plasma ACTH. In all patients with adrenocortical tumours or nodular adrenal hyperplasia (n = 8), ACTH was undetectable (<10 ng l1) [28], a nding which is a clear indication for computed tomography imaging (CT) or magnetic resonance imaging (MRI) scanning of the adrenal glands. Conversely, in all patients with CD, ACTH was detectable, ranging from 12 to 128 ng l1 (NR 1050 ng l1). A corticotrophin-releasing hormone (CRH) test (1 mg kg1 i.v. or 100 mg i.v.) is routinely performed. In all 27 CD patients studied, serum cortisol increased by >20% (range 106554) [53]. Although it is arguable that EAS is so rare in children that the CRH test is not justied, an increased response might contribute to the diagnosis of CD. Recently a routine high dose dexamethasone suppression test (HDDST) has been discontinued after an analysis of cortisol suppression during the LDDST and HDDST [54]. In 24 patients with CD, mean baseline serum cortisol values of 590.7 168.8 nmol l1 decreased to 337.4 104.0 nmol l1 at 48 h during LDDST (P < 0.05); and 66% decreased by >30%. Cortisol suppression during LDDST correlated with that during HDDST (P < 0.05). Consequently, a decrease in cortisol during the LDDST strongly supports the diagnosis of CD. Biochemical features in nodular adrenal hyperplasia The diagnosis of PPNAD has been reviewed recently [23]. In our series, patients had features of primary adrenal CS with raised UFC levels, failure of serum cortisol to suppress on LDDST and HDDST, undetectable plasma ACTH and an absent cortisol response during a CRH test. Some patients with atypical CS tend to have normal or minimally raised 24 h UFC values with absence of a normal cortisol circadian rhythm. In addition, a paradoxical increase of UFC can be demonstrated in patients in the second phase of a HDDST, a feature that can be diagnostic for PPNAD [23]. Genetics of nodular adrenal hyperplasia Sporadic forms of PPNAD without clinical features or family history of CNC are relatively rare [55,56]. Approximately 70% of CNC cases are familial and inherited in an autosomal dominant fashion [25,30,57]. CNC is a genetically heterogeneous condition and linkage studies indicate two predominant genetic loci. The 2p16 locus (CNC2) was identied rst but the gene responsible remains unknown [58]. In recent years, inactivating germline mutations of the regulatory subunit type 1-a of the protein kinase A (PKAR1A) have been identied at the second genetic locus (17q2224) [25]. Mutations are found most frequently in exons 4B, 2 and 6 of the PKAR1A gene resulting in a premature stop codon. To date, PKAR1A gene mutations have been observed in 4050% of families with CNC [59,60] and more recently a third locus has been implicated in at least one large family [30]. There appears to be a higher incidence of PKAR1A mutations if PPNAD is present as part of the CNC clinical picture [24,30].

Box 2. Scheme of investigation for patients with suspected Cushings syndrome

Confirmation or exclusion of Cushings syndrome 1. Urinary free cortisol excretion (24 h urine collection) daily for 3 days 2. Serum cortisol circadian rhythm study (09.00 h, 18.00 h, midnight [sleeping]) 3. Low-dose dexamethasone suppression test (LDDST)  Dose: 0.5 mg 6 hourly (09.00 h, 15.00 h, 21.00 h, 03.00 h) for 48 h  Dose for patients weighing <40 kg; 30 mg kg1 day1  Serum cortisol measured at 0 and 48 h Definition of aetiology of Cushings syndrome 1. Plasma ACTH (09.00 h) 2. CRH test (1.0 mg kg1 i.v.) 3. Analysis of change in serum cortisol during LDDST 4. Adrenal or pituitary MRI scan 5. Bilateral inferior petrosal sinus sampling for ACTH (with CRH)
www.sciencedirect.com

Review

TRENDS in Endocrinology and Metabolism

Vol.18 No.4

171

Radiological investigations Adrenal imaging Adrenal imaging is an essential part of the investigation of primary adrenal CS. The differential diagnosis is between adrenocortical tumour and primary nodular adrenal hyperplasia. Most adrenal tumours are easily visible on adrenal CT or MRI scan, which are the imaging techniques of choice, with higher sensitivity and accuracy than ultrasound scanning [1,61]. In PPNAD, the adrenals are usually of normal size [62]. Although the adrenocortical nodules are small (often <6 mm), occasionally they are visible on CT or less often MRI scanning [63,64]. Pituitary imaging Pituitary imaging (Table 1) using MRI is an important step towards the successful treatment of Cushings disease by trans-sphenoidal surgery (TSS). However, as previously mentioned, most paediatric ACTH-secreting pituitary tumours are microadenomas with a diameter <5 mm [38]. Most of these tumours have a hypointense signal on MRI, which fails to enhance with gadolinium [51]. In the large NIH series, $50% of microadenomas were visible on pituitary MRI [36]. In our series pituitary imaging was of relatively little help, showing a normal appearance in over half of the patients, with a low predictive value of the position of the adenoma, as identied at surgery (Table 1) [39]. Bilateral inferior petrosal sinus sampling for ACTH The technique of bilateral inferior petrosal sinus sampling (BIPSS) was developed mainly at the NIH during the 1980s [65] and has become routine in adult practice. It was hoped that BIPSS would distinguish CD from EAS and also provide a method of identifying a lateral or central source of pituitary ACTH secretion [51]. In children, because of the extreme rarity of EAS, the aim of BIPSS is primarily to demonstrate possible lateralization of ACTH secretion. The rst paediatric data were reported in the large NIH series [36] in which a predictive value of lateralization was 7580% [1]. BIPSS has been performed in paediatric patients in our unit since 1987 and our experience indicates that ACTH sampling gives a better prediction of the site of the microadenoma than does pituitary imaging (Table 2) [66]. BIPSS is a highly specialized technique and should be performed by a radiologist who regularly studies adult patients. General anaesthesia (GA) might alter ACTH secretion. The youngest patient we studied without GA was aged 8.4 years. BIPSS was performed in 24 paediatric CD patients, without complications, and lateralization (interpetrosal sinus ACTH ratio of !1.4 after CRH) [51]

Figure 3. Interpetrosal ACTH ratio during bilateral inferior petrosal sinus sampling in paediatric patients with Cushings disease. A ratio of !1.4 (dotted line) indicates lateralization of ACTH secretion; 19 out of 23 patients (83%) showed lateralization of ACTH secretion. These data are from our unit at the William Harvey Research Institute.

was present in 79% of patients (Figure 3). A more recent report from the NIH described BIPSS in 94 paediatric patients reporting localization of ACTH secretion in agreement with the surgical nding in 58% of cases, thus concluding that the technique was not an essential part of paediatric investigation [67]. The percentage of lateralization, however, increased to 70% (51/73) after exclusion of 18 centrally located and four bilateral lesions. Treatment of Cushings syndrome Primary adrenal lesions First-line therapy for cortisol-secreting adrenocortical tumours is surgical excision. Glucocorticoid replacement is required per- and post-operatively because function of the contralateral adrenal cortex will be suppressed. There is controversy concerning the optimum management of adrenocortical carcinoma with metastases. Mitotane (o,p-DDD) therapy appears the treatment of choice; however, the benet of possible addition of cytotoxic agents has not yet been demonstrated. Further discussion is outside the scope of this review.

Table 1. Pituitary imaging, surgical identication of adenoma and cure by TSSa

Total patients (n) 31
a

Adenoma CT/MRI image (n) 17 (55%)

Concordance of image with surgery (n) 9 (52%)

Cure by TSS (n) 20 (64%)

Abbreviations: CT, computed tomography imaging; MRI, magnetic resonance imaging; n, number of patients; TSS, trans-sphenoidal selective adenomectomy.

Table 2. BSIPSS results, surgical identication of adenoma and cure by TSSa

Total patients (n) 24
a

Lateralization (n) 19 (79%)

Non-lateralization (n) 5 (21%)

Concordance of BSIPSS result with surgery (n) 20 (83%)

Cure by TSS (n) 18 (75%)

Abbreviations: BSIPSS, bilateral simultaneous inferior petrosal sinus sampling; n, number of patients; TSS, trans-sphenoidal selective adenomectomy.

www.sciencedirect.com

172

Review

TRENDS in Endocrinology and Metabolism Vol.18 No.4

The denitive treatment of choice for PPNAD or AIMAH is open or laparoscopic bilateral adrenalectomy [68]. This is not only to treat the CS but also to prevent the secondary complications of hypercortisolaemia and the risk of development of adrenocortical neoplasia. Medical treatment, such as metyrapone, can be used to normalize serum cortisol concentrations before surgery or in patients with subclinical CS. Adrenalectomy for paediatric PPNAD or AIMAH appears safe and effective and post-operatively patients might demonstrate catch-up growth [28]. After cure of PPNAD by adrenalectomy, patients will require long-term corticosteroid replacement and life-long endocrine followup. Patients also require regular clinical and biochemical screening for the associated manifestations of CNC, especially if a mutation in the PKAR1A gene is identied or there is a known family history [25]. Cushings disease CD causes considerable morbidity and requires prompt and expert treatment, which should be curative. The approach to treatment has evolved over the years. Initially, bilateral adrenalectomy was widely practised and, while effective in lowering hypercortisolaemia, the pituitary adenoma remained in situ and there was an appreciable risk of post-adrenalectomy Nelsons syndrome [69,70]. In the management of 32 cases of CD in our unit, adrenalectomy has been performed only twice, when the patients were too unwell to undergo pituitary surgery. In one of these patients, the hypercortisolaemia was uncontrollable by oral metyrapone and treatment was given with i.v. etomidate which successfully controlled the cortisol levels before adrenalectomy [71]. Medical therapy to lower cortisol using metyrapone or ketoconazole is a short-term option but cannot be recommended as a long-term therapy. Trans-sphenoidal surgery TSS consisting of selective removal of the adenoma is now considered rst-line therapy for both adult and paediatric CD. TSS is regarded as a safe and effective procedure in children [7275]. Studies of CD in adult show variable surgical success rates depending on which denition of cure is adopted. Our adult endocrine unit has traditionally used the denition of undetectable (<50 nmol l1) postoperative serum cortisol at 09.00h for successful treatment [76]. The same denition is used for children. In adult practice, recurrence of CD following TSS, using this denition of cure, is extremely uncommon [76]. Selective microadenomectomy can be technically very difcult in children. As discussed above, the microadenomas can be very small and there is an appreciable rate of failure, in terms of denite cure, even in the hands of the most experienced neurosurgeons. Our experience over the past 25 years and the factors that contributed to surgical cure were reported recently [46]. The overall cure rate from TSS in 30 paediatric patients treated from 1982 to 2007 was 64%, while in 24 of these patients, treated since routine BIPSS was introduced during pre-operative preparation, the cure rate was 75% (Table 2). We therefore consider that the ability of BIPSS to identify correctly the lateral or central position of the adenoma has contributed to an increased rate
www.sciencedirect.com

of surgical success [46]. Other paediatric series report cure rates from 45% to 78% [44,45,72,73,77,78], but very few report rates of >90% [36,38]. Successful selective adenomectomy consists of retaining normal pituitary tissue, which is important for the childs development and quality of life. Post-operative hypopituitarism is, therefore, a potential complication of TSS; however, low rates have been reported in several large series [73,77]. An important potential hormone deciency for future growth is that of growth hormone (GH) (see below). Pituitary radiotherapy Pituitary radiotherapy (RT) has been considered a therapeutic option for paediatric CD for many years. Children with CD respond more rapidly to RT than do adults [79,80]. In our centre, external beam RT is used as secondline therapy, after unsuccessful TSS. The decision to proceed to RT can be made usually within 24 weeks of TSS, when it is clear from circulating cortisol concentrations that complete removal of the adenoma has not been achieved [81]. The RT protocol we follow consists of delivering 45 Gy in 25 fractions over 35 days [82]. In 12 patients treated during the past 25 years the cure rate was 92%, which occurred at a mean interval of 0.83 years (range 0.132.86) after completion of therapy. Long-term pituitary function in six of these patients showed that, although GH deciency (GHD) was frequent soon after RT, some recovery occurred in adult life [83]. Gonadotrophin secretion was generally preserved with normal, or early [84], puberty, and TSH and ACTH deciency was minimal. Cognitive function after pituitary RT was not studied, but there are no data that demonstrate a decrease in IQ related to targeted pituitary RT, as opposed to therapy using a broader eld as for paediatric brain tumours. Post-cure growth and development Most patients with paediatric CD have subnormal growth at diagnosis [1,85]. The challenge is to reverse this pattern and achieve acceptable adult height and body composition. A key article from the NIH describes the abnormalities of height and GH secretion [86] together with a rather pessimistic view of post-treatment catch-up growth and adult height [87]. Disappointing post-cure catch-up growth can occur, attributed possibly to continuing GHD, occurring from TSS, pituitary RT or from the long-standing effects of chronic hypercortisolaemia on pituitary and growth plate physiology [85]. GH secretion should be suspected and tested for three months after TSS or completion of RT. If GHD is demonstrated, GH therapy in a standard dose of 0.025 mg/kg/day can be started. GnRH analogue therapy can be added to delay puberty and epiphyseal closure. Results demonstrate that catch-up growth usually occurs and in most patients; adult height within range of target height is achieved [88]. GHD can persist for many years [89], but was usually not sufciently severe to require GH replacement in adult life. Normal body composition is more difcult to achieve. Many patients remain obese and mean BMI SDS was high (P < 0.01) at a mean interval of 3.9 years after cure in 14 patients [88]. A long-term follow-up study of childhood and adolescent CD patients showed that total body fat and the

Review

TRENDS in Endocrinology and Metabolism

Vol.18 No.4

173

ratio of visceral to subcutaneous fat was abnormally high in the majority of patients studied 7 years after cure [90]. Bone mineral density was closer to normal, a nding that we also reported, and some patients had normal bone mineral density at diagnosis [91]. Concluding remarks Paediatric CS shares aetiological origins with adult CS but contains several features that characterize this age range. Early diagnosis remains a challenge because of the lack of appreciation of the nature of the pathology by parents and general practitioners, but demonstration of growth failure is characteristic. Once suspected, the patient requires investigation using a formal protocol and the choice and interpretation of tests is most productively discussed with an adult specialist. CD presents the most difcult challenge in terms of effective therapy. Referral to a centre where paediatric and adult endocrinology, TSS and pituitary RT are available would be optimal. The choice of a neurosurgeon experienced in TSS in children is likely to improve signicantly the chance of cure. Whereas the prognosis for cure is good in most patients with CS, post-treatment management presents challenges for optimization of growth, puberty and body composition.
References
1 Magiakou, M.A. and Chrousos, G.P. (2002) Cushings syndrome in children and adolescents: Current diagnostic and therapeutic strategies. J. Endocrinol. Invest. 25, 181194 2 Savage, M.O. and Besser, G.M. (1996) Cushings disease in childhood. Trends Endocrinol. Metab. 7, 213216 3 Lumbroso, S. et al. (2004) Activating Gsa mutations: analysis of 113 patients with signs of McCune-Albright syndrome a European collaborative study. J. Clin. Endocrinol. Metab. 89, 21072113 4 Yoshimoto, M. et al. (1991) A case of neonatal McCune-Albright syndrome with Cushing syndrome and hyperthyroidism. Acta Paediatr. Scand. 80, 984987 5 Kirk, J.M. et al. (1999) Cushings syndrome caused by nodular adrenal hyperplasia in children with McCune-Albright syndrome. J. Pediatr. 134, 789792 6 Boston, B.A. et al. (1994) Activating mutation in the stimulatory guanine nucleotide-binding protein in an infant with Cushings syndrome and nodular adrenal hyperplasia. J. Clin. Endocrinol. Metab. 79, 890893 7 Mayer, S.K. et al. (1997) Childhood adrenocortical tumours : case series and reevaluation of prognosis a 24-year experience. J. Pediatr. Surg. 32, 911915 8 Stratakis, C. (2000) Cushing syndrome and Addison disease. In Adrenal Disease in Childhood (Hughes, I.A. and Clark, A.J.L., eds), Karger 9 Ribeiro, R.C. et al. (2001) An inherited p53 mutation that contributes in a tissue-specic manner to paediatric adrenal cortical carcinoma. Proc. Natl. Acad. Sci. USA 98, 93309335 10 Michalkiewicz, E. et al. (2004) Clinical and outcome characteristics of children with adrenocortical tumours: a report from the International Paediatric Adrenocortical Tumour Registry. J. Clin. Oncol. 22, 838845 11 Latronico, A.C. et al. (2001) An inherited mutation outside the highly conserved DNA-binding domain of the p53 tumour suppressor protein in children and adults with sporadic adrenocortical tumours. J. Clin. Endocrinol. Metab. 86, 49704973 12 Stratakis, C.A. (2004) Genetics of adrenocortical tumours: gatekeepers, landscapers and conductors in symphony. Trends Endocrinol. Metab. 14, 404410 13 Sbragia, L. et al. (2005) Adrenocortical tumours in Brazilian children. Arch. Pathol. Lab. Med. 129, 11271131 14 Isidori, A.M. et al. (2006) The ectopic adrenocorticotropin syndrome: clinical features, diagnosis, management and long-term follow-up. J. Clin. Endocrinol. Metab. 91, 371377
www.sciencedirect.com

15 Ilias, I. et al. (2005) Cushings syndrome due to ectopic corticotropin secretion: twenty years experience at the National Institutes of Health. J. Clin. Endocrinol. Metab. 90, 49554962 16 Salgado, L.R. et al. (2006) Ectopic ACTH syndrome: our experience with 25 cases. Eur. J. Endocrinol. 155, 725733 17 Hannah, J. et al. (1988) Oncocytic carcinoid of the kidney associated with periodic Cushings syndrome. Cancer 61, 21362140 18 Amano, S. et al. (1978) Ectopic ACTH-MSH producing carcinoid tumour with multiple endocrine hyperplasia in a child. Acta Pathol. Jpn 28, 721730 19 Hsaio, J.C. et al. (1991) Ectopic ACTH syndrome due to clear cell sarcoma of the kidney. Child Nephrol. Urol. 11, 103106 20 Hinnie, J. et al. (2000) Cushings syndrome in a 16 year old girl due to ectopic ACTH precursor production from a pancreatic tumour. Clin. Endocrinol. (Oxf.) 53, 539541 21 Thomas, R.J. et al. (1998) Wilms tumor presenting as Cushings syndrome. Pediatr. Surg. Int. 13, 293294 22 Espinassa-Holder, M. et al. (2000) Paraneoplastic Cushings syndrome due to adrenal neuroblastoma. Med. Pediatr. Oncol. 34, 231233 23 Stratakis, C.A. et al. (1999) Paradoxical response to dexamethasone in the diagnosis of primary pigmented nodular adrenocortical disease. Ann. Intern. Med. 131, 585591 24 Stergiopoulos, S.G. et al. (2003) Human tumors associated with Carney complex and germline PRKAR1A mutations: a protein kinase A disease! FEBS Lett. 546, 5964 25 Stratakis, C.A. et al. (2001) Clinical and molecular features of the Carney complex: diagnostic criteria and recommendations for patient evaluation. J. Clin. Endocrinol. Metab. 86, 40414046 26 Shenoy, B.V. et al. (1984) Bilateral primary pigmented nodular adrenocortical disease. Rare cause of the Cushings syndrome. Am. J. Surg. Pathol. 8, 335344 27 Carney, J.A. et al. (1985) The complex of myxomas, spotty pigmentation and endocrine overactivity. Medicine (Baltimore) 64, 270283 28 Storr, H.L. et al. (2004) Clinical features, diagnosis, treatment and molecular studies in paediatric Cushings syndrome due to primary nodular adrenocortical hyperplasia. Clin. Endocrinol. (Oxf.) 61, 553559 29 Gunther, D.F. et al. (2004) Cyclical Cushings syndrome presenting in infancy: an early form of primary pigmented nodular adrenocortical disease or a new entity? J. Clin. Endocrinol. Metab. 89, 31733182 30 Sandrini, F. and Stratakis, C.A. (2003) Clinical and molecular genetics of Carney complex. Mol. Genet. Metab. 78, 8392 31 Groussin, L. et al. (2002) Molecular analysis of the cyclic AMPdependent protein kinase A (PKA) regulatory subunit 1A (PRKAR1A) gene in patients with Carney complex and primary pigmented nodular adrenocortical disease (PPNAD) reveals novel mutations and clues for pathophysiology: augmented PKA signaling is associated with adrenal tumorigenesis in PPNAD. Am J Hum Genet. 71, 14331442 32 Groussin, L. et al. (2002) Mutations of the PRKAR1A gene in Cushings syndrome due to sporadic primary pigmented nodular adrenocortical disease. J. Clin. Endocrinol. Metab. 87, 43244329 33 Stratakis, C.A. and Kirschner, M. (1998) Clinical and genetic analysis of primary bilateral adrenal diseases (micro- and macronodular disease) leading to Cushings syndrome. Horm. Metab. Res. 30, 456463 34 Carney, J. and Young, W.F. (1992) Primary pigmented nodular adrenocortical disease and its associated conditions. Endocrinologist 2, 621 35 Stratakis, C.A. and Ball, D.W. (2000) A concise genetic and clinical guide to multiple endocrine neoplasia and related syndromes. J. Pediatr. Endocrinol. Metab. 13, 457465 36 Magiakou, M.A. et al. (1994) Cushings syndrome in children and adolescents. Presentation, diagnosis and therapy. N. Engl. J. Med. 331, 629636 37 Weber, A. et al. (1995) Investigation, management and therapeutic outcome in 12 cases of childhood and adolescent Cushings syndrome. Clin. Endocrinol. (Oxf.) 43, 1928 38 Fahlbusch, R. et al. (1994) Neurosurgical management of Cushings disease in children. In Frontiers of Paediatr Neuroendocrinol (Savage, M.O., Bourguignon, J.P. and Grossman, A.B., eds), pp. 6872, Oxford, Blackwell Scientic Publications 39 Storr, H.L. et al. (2005) Factors inuencing cure by transsphenoidal selective adenomectomy in paediatric Cushings disease. Eur. J. Endocrinol. 152, 825833

174

Review

TRENDS in Endocrinology and Metabolism Vol.18 No.4

40 Khadilkar, V.V. et al. (2004) Cushings disease in an 11-month-old child. Indian Pediatr. 41, 274276 41 Damiani, D. et al. (1998) Pituitary macroadenoma and Cushings disease in paediatric patients: patient report and review of the literature. J. Ped. Endocrinol. Metab. 11, 665669 42 Stratakis, C.A. et al. (2000) Pituitary macroadenoma in a 5-year-old: an early expression of multiple endocrine neoplasia type 1. J. Clin. Endocrinol. Metab. 85, 47764780 43 Besser, G.M. and Trainer, P.J. (2002) Cushings syndrome. In Comp Clin Endocrinol (3rd edn) (Besser, G.M. and Thorner, M.O., eds), pp. 193202, Mosby 44 Leinung, M.C. et al. (1995) Long-term follow-up of transsphenoidal surgery for the treatment of Cushings disease in childhood. J. Clin. Endocrinol. Metab. 80, 24752479 45 Devoe, D.J. et al. (1997) Long term outcome in children and adolescents after trans-sphenoidal surgery for Cushings disease. J. Clin. Endocrinol. Metab. 82, 31963202 46 Storr, H.L. et al. (2004) Pre-pubertal Cushings is more common in males, but there is no increase in severity at diagnosis. J. Clin. Endocrinol. Metab. 89, 38183820 47 Greening, J.E. et al. (2006) Linear growth and body mass index in paediatric patients with Cushings disease or simple obesity. J. Endocrinol. Invest. 29, 885887 48 Peters, C.J. et al. Factors inuencing skeletal maturation at diagnosis of paediatric Cushings disease. Horm. Res. (in press) 49 Dupuis, C. et al. Abnormal puberty in paediatric Cushings disease; relationships with adrenal androgens, sex hormone binding globulin and gonadotrophin concentrations. Clin. Endocrinol. (Oxf.) (in press) 50 Hauffa, B.P. et al. (1984) Dissociation between plasma adrenal androgens and cortisol in Cushings disease and ectopic ACTHproducing tumour: elation to adrenarche. Lancet 1, 13731376 51 Newell-Price, J. et al. (1998) The diagnosis and differential diagnosis of Cushings and pseudo-Cushings syndrome. Endocr. Rev. 19, 647672 52 Arnaldi, G. et al. (2003) Diagnosis and complications of Cushings syndrome: a consensus statement. J. Clin. Endocrinol. Metab. 88, 55935602 53 Peters, C.J. et al. (2006) The role of corticotrophin-releasing hormone in the diagnosis of Cushings syndrome. Eur. J. Endocrinol. 155, S93S98 54 Dias, R. et al. (2006) The discriminatory value of the low-dose dexamethasone suppression test in the investigation of paediatric Cushings syndrome. Horm. Res. 65, 159162 55 Mellinger, R. and Smith, R.W. (1956) Studies of the adrenal hyperfunction in 2 patients with atypical Cushings syndrome. J. Clin. Endocrinol. Metab. 16, 350366 56 Cheung, P.S. and Thompson, N.W. (1989) Carneys complex of primary pigmented nodular adrenocortical disease and pigmentous and myxomatous lesions. Surg. Gynecol. Obstet. 168, 413416 57 Akbas, H. et al. (1997) Surgical treatment of left-atrial myxoma in Carneys complex. Thor. Cardiovasc. Surg. 45, 148150 58 Carney, J. et al. (1986) Dominant inheritance of the complex of myxomas, spotty skin pigmentation and endocrine overreactivity. Mayo Clin. Proc. 61, 165172 59 Stratakis, C. et al. (1996) Carney complex, a familial multiple neoplasia and lentiginosis syndrome. Analysis of 11 kindreds and linkage to the short arm of chromosome 11. J. Clin. Invest. 97, 699705 60 Kirschner, L.S. et al. (2000) Genetic heterogeneity and spectrum of mutations of the PRKAR1A gene in patients with the Carney complex. Hum. Mol. Genet. 9, 30373046 61 Hanson, J.A. et al. (1996) Magnetic resonance imaging of adrenocortical adenomas in children: correlation with computed tomography and ultrasound. Pediatr. Radiol. 26, 794799 62 Doppman, J.L. (1997) Problems in endocrinologic imaging. Endocrinol. Metab. Clin. North Am. 26, 973991 63 Doppman, J.L. et al. (1989) Cushing syndrome due to primary pigmented nodular adrenocortical disease: Findings at CT and MR imaging. Radiology 172, 415420 64 Rockall, A.G. et al. (2004) CT and MR imaging of the adrenal glands in ACTH-independent Cushing syndrome. Radiographics 24, 435452 65 Oldeld, E.H. et al. (1991) Petrosal sinus sampling with corticotropin-releasing hormone to distinguish Cushings disease from pseudo-Cushings states or normal physiology. N. Engl. J. Med. 325, 897905
www.sciencedirect.com

66 Lienhardt, A. et al. (2001) Relative contributions of inferior petrosal sinus sampling and pituitary imaging in the investigation of children and adolescents with ACTH-dependent Cushings syndrome. J. Clin. Endocrinol. Metab. 86, 57115714 67 Batista, D. et al. (2006) An assessment of petrosal sinus sampling for localisation of pituitary microadenomas in children with Cushing disease. J. Clin. Endocrinol. Metab. 91, 221224 68 Stratakis, C. (2000) Cushings syndrome and Addison disease. Endocrine Dev. 2, 150173 69 Hopwood, N.J. and Kenny, F.M. (1977) Incidence of Nelsons syndrome after adrenalectomy for Cushings disease in children: results of a nationwide survey. Am. J. Dis. Child. 131, 13531356 70 McArthur, R.G. et al. (1979) Childhood Cushing disease: results of bilateral adrenalectomy. J. Pediatr. 95, 214219 71 Greening, J.E. et al. (2005) Efcient short-term control of hypercortisolaemia by low-dose etomidate in severe paediatric Cushings syndrome. Horm. Res. 64, 140143 72 Massoud, A.F. et al. (1997) Transsphenoidal surgery for pituitary tumours. Arch. Dis. Child. 76, 398404 73 Knappe, U.J. and Ludecke, D.K. (1996) Transnasal microsurgery in children and adolescents with Cushings disease. Neurosurgery 39, 484493 74 Kanter, A.S. et al. (2005) Single-center experience with paediatric Cushings disease. J. Neurosurg. 103 (Suppl. 5), 413420 75 Joshi, S.M. et al. (2005) Cushings disease in children and adolescents: 20 years of experience in a single neurosurgical centre. Neurosurgery 57, 281285 76 Trainer, P.J. et al. (1993) Transsphenoidal resection in Cushings disease : undetectable serum cortisol as the denition of successful treatment. Clin. Endocrinol. (Oxf.) 38, 7378 77 Linglart, A. and Visot, A. (2002) Cushings disease in children and adolescents. Neurochirurgie 48, 271280 78 Styne, D.M. et al. (1984) Treatment of Cushings syndrome in childhood and adolescence by transsphenoidal microadenomectomy. N. Engl. J. Med. 310, 889893 79 Jennings, A.S. et al. (1977) Results of treating childhood Cushings disease with pituitary irradiation. N. Engl. J. Med. 297, 957962 80 Thoren, M. et al. (1986) Treatment of Cushings disease in childhood and adolescence by stereotactic pituitary irradiation. Acta Paediatr. Scand. 75, 388395 81 Storr, H.L. et al. (2003) Clinical and endocrine responses to pituitary radiotherapy in paediatric Cushings disease. J. Clin. Endocrinol. Metab. 88, 3437 82 Plowman, P.N. (1997) Pituitary radiotherapy: techniques and potential complications. In Clinical Endocrine Oncology (Sheaves, R., Jenkins, P.J. and Wass, J.A.H., eds), pp. 185188, Oxford, Blackwell Scientic Publications 83 Chan, L. et al. Long-term anterior pituitary function in patients with paediatric Cushings disease treated with pituitary radiotherapy. Eur. J. Endocrinol. (in press) 84 Nicholl, R.M. et al. (1993) Acceleration of pubertal development following pituitary radiotherapy for Cushings disease. Clin. Oncol. 5, 393394 85 Lebrethon, M-C. et al. (2000) Linear growth and nal height after treatment of Cushings disease in childhood. J. Clin. Endocrinol. Metab. 85, 32623265 86 Magiakou, M.A. et al. (1994) Suppressed spontaneous and stimulated growth hormone secretion in patients with Cushings disease before and after surgical cure. J. Clin. Endocrinol. Metab. 78, 131137 87 Magiakou, M.A. et al. (1994) Final stature in patients with endogenous Cushings syndrome. J. Clin. Endocrinol. Metab. 79, 10821085 88 Davies, J.H. et al. (2005) Final adult height and body mass index after cure of paediatric Cushings disease. Clin. Endocrinol. (Oxf.) 62, 466472 89 Carroll, P.V. et al. (2004) Successful treatment of childhood-onset Cushings disease is associated with persistent reduction in growth hormone secretion. Clin. Endocrinol. (Oxf.) 60, 169174 90 Leong, G.M. et al. (2007) Effects of child- and adolescent-onset endogenous Cushing syndrome on bone mass, body composition and growth: a 7-year prospective study into young adulthood. J. Bone Miner. Res. 22, 110118 91 Scommegna, S. et al. (2005) Bone mineral density at diagnosis and following successful treatment of disease. J. Endocrinol. Invest. 28, 231235