Prader-Willi syndrome: a review of clinical, genetic, and endocrine findings

Severe hypotonia is consistently observed at birth and during the neonatal period [3]; therefore, PWS should be considered in all cases of unexplained neonatal hypotonia. Other features noted during the neonatal period include lethargy, feeding difficulties, thick saliva, and increased head/chest circumference ratio, small genitalia in both males and females with frequent cryptorchidism in males. In older untreated children with obesity, developmental delay, short stature and/or decreased growth velocity, and dysmorphic features are found including a narrow bifrontal diameter, almond-shaped palpebral fissures, a thin upper lip with a down-turned mouth, small hands and feet, straight borders of ulnar side of hands and of inner legs (Fig. 1) [3]. Clinical diagnostic criteria were established by consensus in 1993 [19]. Subsequently, definitive molecular genetic testing became available for laboratory diagnosis of PWS. These clinical criteria were later modified to help define people for whom further diagnostic testing is indicated and at different ages (Table 1) [20]. DNA methylation analysis is the most efficient way to confirm the diagnosis if PWS is suspected clinically but will not identify the genetic subtype [3, 21, 22].

Classically, two nutritional developmental phases have been described in PWS: Phase 1, in which the individual exhibits poor feeding and hypotonia, often with FTT; and Phase 2, which is characterized by “hyperphagia leading to obesity” [3, 6, 13] but recently a total of seven different nutritional phases, with five main phases and sub-phases in Phases 1 and 2 have been identified [23]. Increase in appetite is seen in Phase 2b at age 4.5–8 years, whereas the classical hyperphagia becomes evident during Phase 3 (Table 2).

The last two decades have seen significant increases in the understanding of mechanisms controlling appetitive behavior, body composition, and energy expenditure. Many regions throughout the central nervous system play critical roles in these processes but the hypothalamus, in particular, receives and orchestrates a variety of signals to bring about coordinated changes in energy balance. Ghrelin, a 28 amino acid peptide produced in the stomach, is the only peripheral hormone to transmit satiety signal. Plasma ghrelin level in obese PWS individuals is higher than any other form of obesity and considered as one of the contributing factors for their obesity [24, 25]. A somatostatin analog infusion in 4 adults [26] and long–acting octreotide infusion in 8 adolescents with PWS effectively suppressed ghrelin elevation before meals but not the appetite [27]. Circulating ghrelin levels are elevated in young children with PWS long before the onset of hyperphagia, especially during the early phase of poor appetite and feeding [28]. Based on these studies, it seems unlikely that high ghrelin levels alone are directly responsible for the switch to the hyperphagic nutritional phases in PWS.

Diabetes mellitus type 2 (T2 DM), a metabolic disorder characterized by hyperglycemia in the context of insulin resistance has been reported in 25 % of adult PWS population [29]. Those individuals with T2 DM had a higher past maximum body weight and a greater likelihood of positive family history. Fasting insulin concentrations and homeostasis model assessment insulin resistance index however, are lower in PWS children than in obese control (P < 0.05) and similar to lean control subjects [30]. Not surprisingly, a study of 74 children with PWS at a median age of 10.2 years showed that none had T2 DM and only 4 % had impaired oral glucose tolerance by OGTT [31]. T2 DM should be managed accordingly with special attention to those children on GH treatment with higher risk for insulin resistance. Periodic fasting serum glucose and insulin levels are recommended before and after initiation of GH treatment.

Errors in genomic imprinting are the cause of PWS. It is considered as a phenomenon of epigenetics and modified dependent on the parental sex contributing the genes where epigenetic changes can control expression or gene activity without changing the DNA structure or base pair sequence [32]. The epigenetic process is reversible and occurs during both male and female gametogenesis. The ‘on or off’ activity of gene expression or regulation is usually through DNA methylation at specific bases (e.g., cytosine). Nearly 150 genes in humans are thought to be imprinted and contain CpG-rich differentially methylated DNA regions that correlate with gene allele activity [18]; several located on chromosome 15.

The genes and transcripts located in the 15q11-q13 region can be grouped into four areas delineated by three common deletion breakpoints (Fig. 2): (1) A group of non-imprinted genes, GCP5, CYFIP1, NIPA1, and NIPA2, are located between proximal 15q11-q13 breakpoints BP1 and BP2 and other genes between 15q11-q13 breakpoints BP2 and BP3 (e.g., P gene) are also expressed equally from the paternal and maternal alleles; (2) imprinted genes (paternal expression only) include MKRN3, MAGEL2, NDN, and the bicistronic SNURF-SNRPN gene; (3) the preferentially maternally expressed genes UBE3A and ATP10A (with disturbances in UBE3A causing Angelman syndrome); and (4) non-imprinted genes show evidence of paternally biased expression (e.g., GABRB3) [2, 33, 34]. The bicistronic SNURF‐SNRPN gene in the 15q11-q13 region is involved in mRNA splicing in the brain encoded by exons 4 to 10. Exons 1 through 3 of this complex gene encode a separate protein, which is involved in the genomic imprinting process [2]. Multiple copies of the so-called C/D box small nucleolar RNAs (snoRNAs) or SNORDs (SNORD64, SNORD107, SNORD109A, SNORD115, and SNORD116) are also located in this region and key in the development of PWS with involvement in RNA processing. SNORDs are encoded by a large extended transcript from the complex SNURF-SNRPN gene locus and not translated into protein [35‐38].

Other protein coding genes in the 15q11-q13 region include MKRN3, MAGEL2, and NDN which are imprinted and paternally expressed. They are located proximally to the imprinting center (IC) located at the SNRPN-SNURF gene complex locus and involved in neural development and brain function [2, 4, 5]. The MAGEL2 gene encodes a protein found in the hypothalamus and other brain areas and recently have been reported to be involved in autism spectrum disorder [39]. It appears to function in circadian rhythm, brain structure development, and human reproduction and infertility. The MKRN3 gene generates specific proteins (makorins), which are abundantly expressed in the brain and involved with hormone regulation and precocious puberty [40]. The NDN gene is thought to play a role in axonal outgrowth and expressed in brain regions possibly involved in regulating respiration rate. Two additional genes (PWRN1, PWRN2) are also located close to the NDN gene with the PWRN1 as a possible novel alternative start site for the SNURF-SNRPN gene complex activity [41]. The UBE3A and ATP10C genes are imprinted, maternally expressed, and paternally silent. The UBE3A gene is involved in Angelman syndrome. Several transcripts in the 15q11-q13 region are read in an anti-sense direction and complementary to DNA sequences of other genes in a reverse direction, including the UBE3A anti-sense transcript. Other genes found in the distal non-imprinted area of the 15q11-q13 region include the gamma aminobutyric acid (GABA) receptor subunits (i.e., GABRB3, GABRG3, GABRA5), the P locus for oculocutaneous albinism type 2 (OCA2) and HERC2 [2, 4, 5]. The HERC2 gene encodes an ubiquitin ligase protein, which is expressed in high levels in the fetus and lower expression in the adult brain, testis, ovary, and muscle tissue [10, 42].

Studies have shown that the GABA receptor subunit genes (i.e., GABRB3, GABRA5) are expressed unequally between the paternal and maternal alleles thereby indicating altered allelic expression. Loss of the paternal allele for these genes produces lower than the expected 50 % expression which indicates paternal bias with more expression from the paternal allele than the maternal allele using lymphoblasts carrying different chromosome 15 defects (e.g., deletions, uniparental disomy) [33, 34]. GABA is an important neurotransmitter with inhibitory capability at the brain level. Thus, alterations in gene expression may be associated with appetite, visual perception, and memory changes. The P gene is involved with pigmentary status in individuals with PWS or Angelman syndrome having the typical chromosome 15q deletion. This oculocutaneous albinism type 2 gene when disturbed in individuals with PWS or Angelman syndrome having the typical 15q11-q13 deletion will show hypopigmentation [2, 4, 5, 43].

There are three main genetic mechanisms that result in PWS: paternal 15q11-q13 deletion, maternal uniparental disomy (UPD) 15, and imprinting defects (ID).

Sleep disruption and sleep-disordered breathing have been linked to significant deficits in neurocognitive function, including poor focus, excessive daytime sleepiness, and irritability in both the general population [55] and in individuals with PWS [56]. Initially prompted by the daytime feature of hypersomnolence, many individuals with PWS were identified with polysomnographic features of sleep-disordered breathing, including obstructive, central, and mixed sleep apnea syndromes. Factors including developmental brain abnormalities, craniofacial dysmorphia, hypotonia, obesity, and chest wall deformities have been cited as factors that can contribute to both the presence and severity of sleep-disordered breathing in PWS [57]. Following their metanalysis of the literature on sleep-disordered breathing in PWS, Sedky et al. [57] concluded that while obstructive sleep apneas (OSA) were closely related to obesity in non-PWS children, it was unclear whether body mass index (BMI) played a significant role in increasing OSA risk in PWS children. Central apneas were more common in infants studied with PWS [58]. Symptomatic narcolepsy, with or without cataplexy has also been reported in up to 35.7 % of children with PWS [57].

Hypocretin containing neurons, located in the hypothalamus are thought to play an important role in maintaining wakefulness as well as influencing eating behavior. Congenital dysfunction/developmental failure of the hypocretin system has also been proposed as a potential etiology in this population [59‐61]. In genome wide expression studies, Bittel et al. [62] reported elevated expression patterns of the hypocretin (HCRT) gene in males with PWS, although an early study by Fronczek et al. [63] did not show a significant difference in the total number of hypocretin-containing neurons among PWS patients and age-matched controls, either in adults or infants.

Excessive daytime sleepiness (EDS) is a very common feature in PWS, occurring in 70–100 % of adults with PWS. Most recently, Mass et al. [64] utilized actigraphic scatter-plots to explore the temporal distribution of EDS with severe disruptive behavior and to identify situations where sleepiness was most likely to occur. Lack of structured activities, particularly in the afternoons and evenings were associated with higher rates of EDS.

GH is now commonly used in the management of PWS. The effect of GH during wakefulness has demonstrated improvements in respiratory mechanics and ventilatory responsiveness [65]. More recently, Katz-Solomon et al. [66] performed studies to assess cardiorespiratory control during sleep following the initiation of GH. They demonstrated improved oxygenation and cardiovascular function at 6 months after the initiation of GH in 16 individuals between ages 2–32 months. However, their study also showed that the ventilatory response to 4 % CO₂ and 100 % O₂ was essentially unchanged during quiet sleep suggesting that the previous changes were unrelated to an improvement in chemoreflex-mediated autonomic drive [66].

Longitudinal follow-up studies performed after 2 years of GH therapy reported by Al-Saleh et al. [67] in 15 children between 0.8 and 15.4 years (median 3.7 years) showed a median obstructive apnea/hypopnea index (OAHI) of 0.8 (0.3–6.1) and central apnea index (CAI) of 0.9 (0.2–2.3). GH was discontinued in two of these children due to the development of severe OSA identified by polysomnography at 6 weeks post-initiation of GH therapy. However, at 2 years post-treatment no significant changes in sleep-disordered breathing were identified in these children, suggesting a period of increased vulnerability in the first few weeks after the initiation of treatment [67].

Berini et al. [68] evaluated adenotonsillar size in 50 children with PWS before the initiation of GH treatment at 6 weeks, at 6 months, at 12 months, and then yearly up to 4 years. Three children developed severe OSA requiring discontinuation of growth hormone therapy. This group found a direct correlation of OAHI with adenoid size but neither with tonsillar size nor with plasma IGF-1 levels [68]. Thus, it is appropriate to screen all individuals with PWS for sleep-disordered breathing. While screening questionnaires and physical examinations may be helpful, neither has shown good specificity or sensitivity [69]. Opinions vary as to the timing and frequency of sleep evaluations but most agree that it is appropriate to study children before the initiation of GH and then periodically, especially with significant changes in weight, prior to and following adeno-tonsillectomy and before spinal or craniofacial surgery.

Whitman and Myers [70] also recommended a repeat polysomnogram at 3–6 months after the initiation of GH in patients with PWS previously diagnosed and treated for OSA as well as regular screening with the Chervin sleep questionnaire (questions 1–6), examination of the oropharynx for tonsillar hypertrophy and monitoring of IGF-1 levels during the entirety of GH therapy. Multiple sleep latency testing (MSLT) is also useful in assessing pathologic sleepiness and is done following an overnight polysomnogram.

In children, adeno-tonsillectomy is the preferred treatment for obstructive sleep apnea. High-risk children including those with PWS and especially those with severe apnea may have residual apneic events and require post-operative evaluation to assess if additional therapies, such as Continuous Positive Airway Pressure (CPAP) are needed [71, 72]. Velopharyngeal dysfunction with hypernasality requiring additional surgical intervention has also been reported following adeno-tonsillectomy in children with PWS [73]. In adults with PWS, CPAP is recommended as the initial therapy due to OSA [74].

Short stature is a main characteristic of individuals with PWS. Children with PWS fail to show the growth acceleration seen in puberty and the mean final height without treatment is 148 cm in girls and 155 in boys [75, 76]. Growth charts for non-GH-treated infants and children with PWS have been developed [77, 78]. Non-PWS obese children have decreased GH secretion while maintaining normal serum insulin-like growth factor-1 (IGF-1) levels and normal height. Children with PWS have short stature with low serum GH and IGF-1 levels, therefore true GHD [76, 79‐81] is considered.

Children with PWS treated with GH through childhood are able to achieve normal adult height [75, 82]. Beneficial effects on body weight, body composition, and exercise capacity found in these studies prompted studies looking at the effects of GH unrelated to height. In one study, 60 prepubertal children ages 3.13–7.16 years had normal height and positive long-term effects on maintaining body composition after 8 years of GH treatment without any adverse effect on glucose homeostasis, serum lipids, blood pressure, and bone maturation [83].

Anabolic effects of GH including increase in lean body mass, motor strength, and decrease in fat mass were reported in 21 children when started at 13 ± 6 months of age for a period of 6 years and compared with 27 untreated children of similar age [80]. A recent study reported low BMI in 61.5 % (33.8 % osteoporosis and 27.7 % osteopenia) of 101 patients with a mean age of 5.4 years (range 3–17 years.) and significant improvement after GH treatment for a mean period of 54 months (range 6–144 months) [84].

The beneficial effect of GH therapy in childhood into adulthood is unclear after GH is discontinued. A report of GH treatment discontinuation after 12 and 24 months in 14 individuals with PWS revealed an increase in BMI-SDS (P = 0.008 and P = 0.003) and visceral fat (P = 0.062 and P = 0.125), respectively [85]. In one study, however, improved mean BMI (32.4 vs. 41.2), body composition, lower mean hemoglobin A1c, lower mean insulin resistance, and less hypertension were reported in 20 adults (mean age 25.4 years) with PWS at 7.0 ± 4.4 years after discontinuing treatment initiated at age 11.8 ± 2.7 years when compared with 40 untreated PWS adult group [86]. Butler et al. [87] studied 11 adults with PWS (average age = 32 years) over a 2-year period with GH treatment during the first year only. Total lean muscle mass and moderate-vigorous physical activity and plasma IGF-1 and high density lipid (HDL) levels were significantly increased while on GH, while percent body fat decreased during the 12 months of GH treatment. IGF-1 and HDL levels returned to near baseline and body fat increased after GH treatment during the second year. In addition to improved body composition, increase in muscle strength and exercise tolerance has been reported after 12 and 24 months of GH treatment in 15 obese adults with PWS [88]. Cardiovascular features in obese PWS adults including smaller left ventricular size and lower systolic function are similar to those reported in adult GHD [89]. In one study, left ventricular mass increased significantly after 1 and 2 years of GH treatment (0.40 ± 0.11 to 0.97 ± 0.17 mg/days) in 9 adult PWS individuals without evident abnormalities of systolic and diastolic function [90]. These data indicate that beneficial effects of GH are still present even after the epiphyses are fused and long-term GH treatment, in addition to strict diet and exercise program may be necessary to maintain good body composition.

Beneficial cognitive effects have been reported in children [91‐93] and adults [94] with PWS during GH treatment. Recent study in 19 children with PWS at median age 6.3 years, showed no changes in cognition and behavior over one or two years of GH treatment [95]. These children however, had marked deterioration in behavior at 6 months after abrupt GH discontinuation. These studies demonstrated that GH maintenance therapy may prevent or slow down the progression of behavioral problems in PWS individuals.

Recombinant human growth hormone (hGH) was FDA approved in the United States in 2000 for the indication of short stature and growth failure due to PWS. In Europe, growth retardation is not required whereas improvement of body composition is included in the approved indication of GH therapy in PWS [96]. Recommended starting dose in children is 0.18–0.3 mg/kg/week given as daily subcutaneous injections with careful monitoring of clinical status, bone age, and serum IGF-1 levels at regular intervals. Potential concerns related to excessively high IGF-1 levels include lymphoid hyperplasia leading to OSA and a theoretical increase in malignancy risk. In one study, IGF-1 and IGF-binding protein-3 (IGFBP3) levels were evaluated over a 2-year period in a group of 33 children with PWS treated with GH. These subjects were compared to 591 subjects treated for GH deficiency. The PWS group had significantly higher IGF-1 levels despite lower doses of GH. However, there was no significant difference in IGF-1 to IGFBP3 molar ratios between the groups, suggesting that bioavailable IGF-1, and therefore risk for adverse effects, may be similar in both groups [97]. In view of existing information, GH dose based on ideal rather than actual body weight and monitoring serum IGF-1 levels between 0 and +2 SDS and/or IGF-1/IGFBP3 molar ratio without exceeding GH pretreatment should be considered to prevent abnormal serum IGF-1 elevation, insulin resistance, and occasional acromegaloid features seen in children with PWS at standard recommended GH dose.

The prevalence of scoliosis in PWS is high (30–80 %) [98, 99]. Scoliosis is a major concern for patients with PWS treated with GH. Prevalence, onset, and progression of scoliosis are not affected by the genotype or by growth hormone treatment [75, 100‐103]. Therefore, scoliosis is no longer a contraindication for GH treatment in children with PWS. However, due to the high prevalence of scoliosis and the potentially associated morbidities in patients with PWS, regular physical examinations and periodic radiographic evaluations of the spine are recommended.

Significant attention has been given to the association of GH therapy and sudden death in PWS. In a review of 64 cases of death in children from a few days to 19 years of age, 28 subjects (44 %) received GH treatment [104]. Respiratory disorders were the most common cause of death among treated and untreated patients. In this review, most of the deaths in GH-treated children (75 %) occurred during the first 9 months after the initiation of GH treatment. High mortality rate and increased risk of sudden infant death have been reported in children with PWS independent of GH therapy and without evidence of an association between death and GH treatment [105, 106]. In view of these data, the initiation of GH is recommended at the lower dose, 0.18 mg/kg/week in infants with PWS.

Benefits have been reported with increased lean body mass and decreased body fat mass after 6–12 months GH treatment in adults [107‐110]; however, GH is not FDA approved for adult PWS individuals unless confirmed by standard adult GH stimulation testing. The prevalence of GH deficiency in adults with PWS ranges from 15 to 95 %, depending on the agents used for stimulation testing and the threshold GH level used to define deficiency [107, 108]. A normal IGF-1 level does not exclude the diagnosis of GHD and provocative testing is mandatory to make the diagnosis of adult GH deficiency (AGHD). The Endocrine Society (ES) recommends that the insulin tolerance test (ITT) and the GHRH- arginine test are sufficiently sensitive and specific to establish the diagnosis of AGHD. Glucagon stimulation test can also be used when GHRH is not available and performance of an ITT is either contraindicated or not practical in a given patient [111]. It is recommended that GH dosing regimens be individualized rather than weight-based and to start with low doses (0.1–0.2 mg) then titrated according to clinical response, side effects, and IGF-1 levels.

The term “congenital hypogonadism” refers to complete or partial pubertal failure due to insufficient secretion of the pituitary gonadotropins LH and FSH and gonadal sex steroids. Hypogonadism represents a common clinical feature in PWS. Clitoral and labia minora hypoplasia in females and micropenis with hypoplastic scrotal sac in males are evident at birth. Unilateral or bilateral cryptorchidism is present in 80–90 % of males [5]. Similar to many other manifestations of PWS, hypogonadism has been classically thought to be hypothalamic in etiology. However, recent evidence has emerged supporting primary gonadal failure as a significant contributor to male hypogonadism [112‐114]. Other studies have also shown a combined picture of hypogonadotropic hypogonadism with relatively low LH levels, and primary hypogonadism with low inhibin B and relatively high FSH levels [114, 115].

The transient increase in gonadotropins and gonadal hormonal levels occurs during the first months of life, and “minipuberty” is normal in infants with PWS [116, 117]. Gonadotropins and testosterone play important roles in testicular descend. If hypogonadotropic hypogonadism is present in boys, decreased LH, FSH, testosterone, and inhibin B levels are found in addition to a micropenis (stretched penis <2.5 cm) and cryptorchidism. These findings suggest that additional factors may be responsible for the high incidence of cryptorchidism in infants with PWS.

Gonadal function has also been evaluated longitudinally in 61 girls with PWS. LH levels were found to be relatively low for the low estradiol levels observed, and FSH levels were normal. Although the girls had a normal onset of puberty, the progression was delayed in comparison to the normal population [118]. The pattern of gonadal dysfunction in PWS females (primary and combination of a primary gonadal defect and hypothalamic dysfunction) seems to be similar to those observed in boys, male adolescents, and adults with PWS [119, 120]. Four females with PWS diagnosed as children (3 with deletions and one with UPD) with onset of menarche in their 20 s became pregnant and delivered children by caesarian section. Two of these mothers had normal children and the two PWS mothers with 15q11-q13 deletions delivered a male and a female with Angelman syndrome also with the deletion [121, 122].

Due to high prevalence of premature adrenarche in PWS [123‐125], commonly associated with pubic or axillar hair, careful assessment, preferably using a Prader orchidometer, is necessary to demonstrate further testicular enlargement as a first sign of puberty in males (testicular volume greater than 4 ml indicates onset of puberty) and breast development Tanner stage II as the first sign of puberty in females with PWS. During adolescence, the laboratory diagnosis of hypogonadotropic hypogonadism is relatively simple with the identification of very low circulating total testosterone and low to low-normal gonadotropin and inhibin B levels. This hormone profile rules out a primary testicular disorder in which testosterone and inhibin B levels are low, whereas FSH and LH are elevated. Serum FSH, LH, estrogen, and inhibin B profiles are recommended in females with PWS before considering sex hormone replacement therapy (SHRT).

The beneficial effect of sex steroids on the muscle mass and bone health in an individual is well known. Previous studies have suggested that sex steroid deficiency contributes to low bone density in adults with PWS [126, 127]. However, no consensus statement exists as to the most appropriate regimen of sex hormone treatment in PWS. The choice of a particular hormone replacement therapy protocol will depend on the age at diagnosis and local practices. In a recent study, the administration of human chorionic gonadotropin (hCG) 250–500 IU intramuscular biweekly injections for 6 weeks in 16 infants (median age 1.6 years) with PWS with undescended testes resulted in an anatomical lower testis position in most and 23 % had complete scrotal descent. Orchidopexy was required in 76 % of cases in order to ensure a stable position in the scrotum [128]. Furthermore, the American Academy of Pediatrics, Committee on Genetics recommends a therapeutic trial of hCG before considering surgery for undescended testes [129]. The benefits of this modality of treatment include avoidance of general anesthesia, increased scrotal sac size and phallus length to facilitate circumcision/micturition, and thereby improving surgical outcomes for undescended testes.

Testosterone replacement treatment can also lead to improvement in quality of life in males with hypogonadism [130]. In general practice, injectable testosterone is preferred for reasons of convenience and cost. Testosterone enanthate (TE), one of the preparations available on the international market, can be injected once every 2 or 3 weeks in adult males with hypogonadism [131]. It has been thought that TE may increase aggressiveness, but this has not been clearly demonstrated. However, it seems reasonable to start as low as 25 % of the recommended normal adult TE dose (200–250 mg) with gradual increase as tolerated to keep low-normal serum testosterone levels in males while under an endocrinologist’s guidance. Virilization of male patients can also be achieved by percutaneous testosterone administration in gel or patch form. However, these alternatives are expensive and require daily administration, raising problems of adherence and the risk for skin irritation and skin picking behavior in PWS.

Guidelines for hormonal replacement therapy in females with PWS are tailored individually depending on sexual development, hormonal profiles, bone density, and emotional and social needs. Oral estrogen alone or in combinations with progestin is well tolerated. Girls with PWS have normal or near normal secondary sexual characteristics including breast development and menses may occur indicating a lesser degree of hypothalamic-gonadal dysfunction in females and therefore counseling about the risk of pregnancy and discussion of birth control during reproductive age would be advisable.

Similar to other endocrinopathies in PWS, the etiology of hypothyroidism is thought to be central in origin. Hypothyroidism has been reported in approximately 20–30 % of children with PWS [132]. Yet, in one study of 47 individuals with PWS age 10–44 years, the prevalence of hypothyroidism was 2.1 % and not different than the normal population of 2 % [133]. Based on low total (T4) and free thyroxine (FT4) in the presence of normal thyroid-stimulating hormone (TSH), one study reported a prevalence of 72.2 % of central hypothyroidism in children with PWS less than 2 years of age [134]. A recent study revealed normal newborn screening in 23 neonates with PWS for congenital hypothyroidism. In the same study, TSH response to thyroid-releasing hormone (TRH), T4 and FT4 in 21 children from birth to adolescence showed normal patterns except in one elder child with central hypothyroidism [135]. Based on these studies, the prevalence of hypothyroidism is variable and cannot be clearly established. Thus, levothyroxine treatment should not be routinely prescribed in children with PWS unless confirmed by thyroid function testing. It is recommended that baseline thyroid function testing (FT4 and TSH) be done during the first 3 months of life unless they have had a normal newborn screening and annually thereafter, especially if the patient is receiving GH therapy.

Based on generalized hypothalamic dysfunction, children and adults with PWS are at risk for central adrenal insufficiency (CAI). The first published cross-sectional analysis in 2008 of adrenal insufficiency in children with PWS, reported CAI in 60 % of cases after receiving an overnight single-dose of metyrapone [136]. However, five subsequent studies using different methodologies, including low- and high-dose Synacthen and insulin tolerance testing did not confirm the reported high frequency of CAI [137‐141].

Isolated ACTH insufficiency is rare but it may be a component of a multiple hormone deficiency (MPHD). Development of additional pituitary hormone deficiencies in children with acquired MPHD or initially diagnosed with growth hormone deficiency (GHD) is characterized by a sequential order of hormonal loss, usually GH, TSH, LH/FSH, and ACTH [142]. In adults, LH/FSH disturbances have been reported after GH problems, then followed by TSH and with ACTH being the last pituitary hormone deficiency [143]. Although additional inherent and exogenous factors may regulate adrenal androgen production, normal ACTH secretion action is needed for adrenarche [144‐146], which is increased in children with PWS. The true prevalence of CAI in PWS therefore remains unclear with no consensus present among endocrinologists as to whether evaluation should be done for CAI and/or glucocorticoid treatment preoperative or during significant stress.

Metyrapone (30 mg/kg orally at midnight) inhibits 11-hydroxylase, the final step in cortisol synthesis, thereby decreasing cortisol secretion and subsequently increasing ACTH secretion. Using metyrapone is a cumbersome test that is rarely performed because of the difficulty in obtaining metyrapone and the risk of precipitating an adrenal crisis. Hence, parents of children with PWS should be counseled about symptoms consistent with adrenal insufficiency and physicians alerted to the risk and treated accordingly should symptoms occur. In summary, PWS is a classic example of a genetic disorder requiring a complex multi-discipline approach to treat ongoing growth, medical, and endocrine disturbances common to all individuals throughout the natural history of this condition.