Introduction
Adult height shows distinct variability in the general population, following a normal Gaussian distribution dependent on age, sex, ethnicity, as well as many other factors. Human growth leading to final height is a composite and dynamic process, associated with phenotypic changes in stature, body proportions, and composition, reflecting the interplay of genetic, environmental, hormonal, nutritional, and socio-economic factors [
1].
Different endocrine factors regulate growth during each period of life, determining separate but closely integrated phases whereby many hormones influence transient growth and final height [
2]. During the intrauterine phase, fetal growth is critically dependent on insulin and Insulin-like Growth Factors (IGF), both maternal and placental, and nutritional status [
3]. In the early years of life, nutrition is a pivotal factor, while in childhood, a crucial role is played by the GH/IGF1 axis. Nevertheless, thyroid hormones, adrenal androgens, glucocorticoids, sex steroids, ghrelin, leptin, and insulin are all known to participate in the growth process through their interactions with the GH/IGF1 axis. During puberty, the growth spurt depends on the interaction between the somatotroph and gonadal axes, which act synergistically towards the achievement of final stature [
4,
5].
In a clinical context, deviations from a normal growth pattern can often represent the first evidence of a huge spectrum of diseases, encompassing both endocrine and non-endocrine disorders [
6,
7]. While these deviations often manifest as growth inhibition or delay, excessive growth leading to tall stature may also reflect underlying pathological alterations.
Traditionally, ‘tall stature’ in children is defined as a height exceeding the 97.7
th percentile or two standard deviations (SD) above the mean height for a population of the same age, sex, and ethnicity [
8]. As a result, 2.3% of children fall into the category of tall stature and thus may be considered worthy of investigation [
9]. Furthermore, children presenting with height within the normal range, but with a projected height surpassing 2SD above their mid-parental height, may also be evaluated. Although the percentage of children with tall stature is equal to that of children with ‘short stature’, referrals to paediatric endocrinologists for the assessment of tall stature appear to be far less frequent than those for short stature. This is probably due to a better societal acceptance of tall stature, along with the assumption that increased height in a child with tall parents is not alarming and is simply an acceptable familial trait. Indeed, clinical referrals become more likely whenever height exceeds > 2.5SD or > 3SD (extremely tall stature), respectively, 0.6% and 0.1% of the population [
1]. Nevertheless, such patients require intensive investigation to establish any underlying pathological cause of increased growth, and to address potential problems of social adaptation [
10].
‘Familial tall stature’, also known as constitutional tall stature, represents the most common cause and is considered as a variant of the normal pattern of childhood growth and development [
11]. However, despite their rarity, many pathologic conditions also present with tall stature and may be associated with severe comorbidities [
12]. Therefore, differentiating between healthy tall children and those with underlying diseases, while ruling out chromosomal, genetic, and endocrine disorders in the latter [
13], poses a significant clinical challenge.
In the assessment of such children, measurement of current height, growth velocity, weight, head circumference, and body mass index (BMI) should be evaluated [
14]. Additional diagnostic information should be gathered from birth data (weight, length, and head circumference), family history (height and pubertal timing of both parents), developmental history, and growth curve review, if available [
15]. Assessment of body proportions is also critical: specifically, a head circumference > + 2SD associated with tall stature could help clinicians to identify overgrowth syndromes, such as the Beckwith-Wiedemann, Sotos, Perlman, Simpson–Golabi–Behmel, Tatton–Brown–Rahman and Weaver syndromes [
16,
17]. These overgrowth syndromes typically affect childhood from the prenatal to the postnatal phases, involving the development of the patients, in particular Beckwith-Wiedemann and Sotos syndromes that may be associated with hormone imbalance and increased susceptibility to malignancy [
17]. The presence of cardiovascular abnormalities, skin anomalies, skeletal malformations, facial dysmorphisms, abnormalities of the genitalia, and neurodevelopmental delay, may also suggest an underlying syndromic cause [
18].
However, many of these genetic conditions exhibit overlapping phenotypes, thus complicating the differential diagnosis, particularly when dealing with patients in transition or adult age. Similar to children with short stature, ‘constitutionally tall’ individuals referred to the endocrinologist for an evaluation of the GH/IGF1 axis typically show no clear biochemical abnormalities [
19]. Indeed, ‘pituitary gigantism’, excessive stature due to a primary hypothalamic-pituitary abnormality, is an extremely rare disease, with an estimated annual incidence of 8 per million, with only several hundred cases reported to date [
20]. It may be a sporadic and isolated condition, and approximately half of the cases occur within the context of a concurrent hereditary syndrome or follow a familial inheritance pattern [
3]. Excessive GH levels not only cause dramatic linear growth acceleration but can also lead to mild to moderate obesity, progressive macrocephaly, prognathism, and changes in glucose metabolism, including type 2 diabetes [
21]. It is therefore vital, given the relevant number of comorbidities connected to a prolonged diagnostic delay, to identify such patients early and initiate appropriate therapy [
22].
Although most children with short or tall stature do not have an underlying pathological condition, extreme deviations from average height, especially beyond + 3SDs, require further investigation. This review aims to provide a practical clinical approach to identify patients in the transition age (15–25 years) who may have underlying hypothalamic-pituitary defects, as opposed to idiopathic/constitutional tall stature. We conducted a comprehensive search of English-language original articles in the MEDLINE (PubMed) database between December 2021 and March 2022. The search used free text words in combination with Medical Subject Heading (MeSH) terms. The keywords applied for the search included the “Gigantism” term as a keyword and “human” as a filter.
Clinical presentation of gigantism during the transition age
Gigantism and acromegaly represent two clinical manifestations of the same pathological entity—namely a GH-secreting pituitary adenoma, also known as a pituitary neuroendocrine tumour (PitNET). The clinical phenotype largely depends on the timing of disease onset in relation to skeletal maturation. GH excess determines a continuum of clinical manifestations that can occur both before and after the fusion of the epiphyseal growth plates, with frequent overlap [
40]: many of these patients will have features of acromegaly in conjunction with gigantism, hence the term ‘acro-gigantism’.
Apart from scattered case reports, only two studies have reported the clinical presentation of patients with pituitary gigantism diagnosed during the transition age [
39,
41,
42]: a single-centre study by Colao et al. conducted on 13 patients diagnosed between 15 and 20 years, and a multi-centre study by Rostomyan et al. investigating more than 200 patients with a wider age at diagnosis (median 21 years, interquartile range 15.5–27) [
39]. Tall stature is usually the first clinical sign that leads to medical attention, thus initiating the diagnostic process for gigantism. In particular, patients with pituitary gigantism show a peculiar growth pattern, in which the young patient, who was not initially born large for gestational age, progressively crosses higher percentiles during childhood, eventually reaching an adult height above 2SD and surpassing their genetic TH [
9]. The onset of growth acceleration has been demonstrated to occur significantly earlier in females than in males (median age of onset: 11
vs 13 years). Additionally, a shorter diagnostic delay from symptoms onset to diagnosis has been found in females, resulting in a significantly lower age at the time of gigantism in females than in males (median age at diagnosis: 15.8
vs 21.5 years) [
39]. Therefore, males are more likely to be diagnosed during the transition age. Of note, not all patients reportedly had attained their final height at the time of diagnosis, particularly male patients [
39]. While tall stature is generally the primary presentation-presenting feature, the pathological effects of the prolonged exposition to supraphysiological levels of GH and IGF1 are systemic. Patients with gigantism may also show acral enlargement and facial changes, which represent the second most frequent clinical sign (37%). The median shoe size reported at diagnosis was 48 (EU) in males and 42 in females. Acromegalic features were already present at diagnosis in patients with gigantism regardless of sex and age, although facial changes were less commonly observed in patients aged < 19 years [
39]; similarly, signs and symptoms typical of acromegaly such as joint disorders and sweating were rarely encountered in younger patients.
Since most patients with gigantism often harbour macro- and giant pituitary tumours, signs and symptoms of compression are frequently observed at diagnosis, with headache and visual field defects being reported in 23% and 12% of patients, respectively. In addition to visual field impairment, lachrymation, transitory eyelid palsy, or ptosis have also been reported [
41]. Furthermore, around a quarter of patients exhibited at least one pituitary deficit; hypogonadism was diagnosed in 40% of patients at diagnosis [
39]. In line with these findings, one study reported the presence of amenorrhoea, both primary and secondary, in all female patients with gigantism [
41]. Prolactin co-secretion has been reported in more than 30% of cases, particularly in patients with invasive and extrasellar pituitary tumours, with galactorrhoea reportedly being slightly more frequent in females [
39]. Moreover, typical GH excess complications such as sleep apnoea, carpal tunnel syndrome, hypertension, and glucose metabolism disorders were already present at diagnosis, particularly in patients aged > 20 years [
39].
Alterations in glucose and lipid metabolism may be seen in patients with gigantism; insulin-resistance has been mainly found at diagnosis [
41], whereas glucose intolerance and overt diabetes mellitus have been reported less frequently at diagnosis, particularly in patients < 19 years [
39].
Concerning cardiovascular disease, cardiac impairment was detected at diagnosis in 36.5% of cases, primarily involving left ventricular hypertrophy (21%) and diastolic dysfunction (10%) [
39]. In one study comparing the echocardiographic parameters of six males diagnosed with pituitary gigantism during adulthood with those of six age- and sex-matched acromegalic patients and ten healthy controls, both groups of patients displayed significantly higher left ventricular mass index, interventricular septum diastolic thickness, and posterior wall thickness compared to controls. Although patients with gigantism exhibited a significantly longer disease duration, no relevant differences in cardiac structure and performance were noted in these patients when compared to acromegalic patients. However, individuals with cardiac abnormalities in the gigantism group exhibited higher IGF1 levels than those with a normal cardiac structure. For this reason, the authors of this paper suggest performing echocardiography regardless of disease duration to detect cardiac impairment early [
40]. Apart from one case [
40], no alterations in blood pressure or heart rate have been found in patients with gigantism [
41].
Thus, in patients diagnosed during the transition age, it is necessary to not only focus on tall stature and external changes but also investigate potential systemic complications.
Clinical approach
The clinical approach to tall children during transition age should include, wherever possible:
-
Birth data (weight, length, and head circumference);
-
Familial auxological parameters: height/weight for parents and first-degree relatives, pubertal timing of parents (age of menarche of the mother, age of the pubertal growth spurt of the father);
-
Personal medical history: hypo/hyper-glycaemia, metabolic disorders, over-feeding, cardiac defects, ocular defects, anosmia, ligamentous laxity, joint dislocation, obesity, and neurodevelopmental disorders;
-
Assessment of standing height, sitting height, arm span, weight, BMI and head circumference, as compared to country-specific growth charts;
-
Assessment of pubertal status according to Prader’s scale, Marshall and Tanner staging;
-
Assessment of HV: calculated at least every 6 months, expressed in cm per year (cm/yr) with particular attention to peak-height-velocity indicating a pubertal spurt [
43]
;
-
Clinical evaluation: cardiac murmurs, anomalies of the skin, skeletal examination (pectus excavatum, scoliosis), and facial dysmorphism;
The initial approach to a tall child in the transition age should also incorporate the determination of bone age according to a standardised model (for example, the Greulich and Pyle atlas or the Tanner-Whitehouse atlas version 2 or 3 [
44]) to distinguish between a physiological constitutional growth delay or familial tall stature—characterised by a normal/delayed bone age—and pathological precocious puberty, characterised by advanced bone age. Especially during the peri-pubertal transition age, it is crucial that growth assessments be performed regularly; when determining the normality of a child's growth pattern, serial height measurements of HV calculations are more useful than a single height-for-age percentile. A child that grows regularly on a high percentile (even above the 97
th percentile), without significant comorbidities, and especially with a family history of tall stature, should generally be considered a normal variant. Conversely, rapid acceleration of growth, regardless of the percentile, should be investigated further to rule out pathological causes [
45]. Transient tall stature can also be observed in patients with true precocious and pseudo-precocious puberty.
The most prevalent cause of tall stature is familial tall stature, characterised by tall parents, normal growth velocity, normal findings on physical examination, and correspondence between bone age and the chronologic age. Stature generally remains in the target genetic range [
8]. Sometimes this condition is characterised by an acceleration in growth velocity in early childhood, between 2 and 4 years of age. Growth progression remains slightly above the normal curve, following the same centiles until puberty. Rarely, children may also exhibit advanced bone age and early pubertal development within the normal range. Pagani et al
. suggested the possibility of GH hypersecretion in children with familial tall stature, as supported by the presence of age- and sex-adjusted IGF1 levels in the upper range of normal, or hypersensitivity to GH [
46].
Another physiological cause of tall stature is, paradoxically, constitutional delay of growth and development (CDGD). This may occur not only in children from short or normal-statured families but also in children of tall-stature families. A study published in 2005 analysed a cohort of adolescents aged 12–16 years and demonstrated that the final height of CDGD children exceeded the mean TH by more than 4 cm, reaching a mean value of + 1.9 and 2.1SD for boys and girls, respectively—consistent with final tall stature—in 42% of cases [
47].
Certain genetic conditions, such as Marfan Syndrome, may be characterised by tall stature, with a rapid increase in growth velocity occurring just before or in the early stages of the transition age. For this reason, Disease-specific Growth Charts of Marfan Syndrome patients have been developed in some countries. The syndrome, caused by mutations in the fibrillin-1 (
FBN1, chromosome 15q) gene and dysregulation of transforming growth factor β (TGFβ), affects the skeletal system, resulting in tall stature, abnormally long and slender limbs, fingers, and toes, chest wall abnormality, and scoliosis. The arm span is greater than their height, with an arm span-to-height ratio greater than 1, while the upper/lower segment ratio is diminished [
48]. A Korean study showed that the 50th percentile of height in patients with Marfan Syndrome exceeds the normative 97
th percentile for both genders [
49]. A French study comparing more than 250 Marfan patients to a control population demonstrated that Marfan children’s overgrowth decreases with age, especially during the transition phase, at about 17 years of age [
50]. Another important aspect deals with the specific mutation of Marfan Syndrome since patients carrying
TGFBR2 mutations have lower mean height than patients harbouring
FBN1 mutations [
51].
Klinefelter syndrome also exhibits its peculiar growth pattern, with normal auxological parameters during infancy, followed by a rapid growth tall stature between 5 and 8 years of age, and further growth in the pubertal period [
52]. Thus, disease-specific growth charts can be useful for monitoring growth patterns, planning the timing of growth-reductive therapy if necessary, and predicting adult height.
In conclusion, for correct identification of tall stature, it is necessary to report the height value on country-specific or disease-specific growth charts. A systematic clinical approach, along with the periodic monitoring of auxological parameters and HV, is essential for distinguishing between physiological and pathological causes of tall stature.
Table
1 summarises the main differential diagnoses for tall stature.
Table 1
Main differential diagnoses for tall stature
Transition Age |
Familial Tall Stature | ↑/ → | < 2 SD | → | N | Normal appearance |
FIPA | NA | > 2 SD | NA | N | Tall stature |
MAS | ↑ | NA | ↑ | P | Café-au-lait spots, skeletal lesions (fibrous osteodysplasia), craniofacial dysplasia (optic and auditory nerve impairment) |
Marfan syndrome | NA | > 97th percentile | ↑ | N | Abnormally long and slender limbs, fingers, toes, chest wall, and scoliosis |
Klinefelter syndrome | → | > 2 SD | ↑/ → * | N | Small, firm testes; gynecomastia; high-pitched voice; learning disability |
Hyperthyroidism | ↑ | < 2 SD | ↑ | N | Goiter, tachycardia, hypertension, diarrhea, exophthalmos |
Obesity | ↑ | > 2 SD | ↑ | P | BMI > 95th percentile |
Childhood |
Beckwith-Wiedemann syndrome | ↑ | ≥ 2 SD | ↓* | NA | Macroglossia, abdominal wall defects, congenital heart disease, |
Sotos syndrome | ↑ | > 2 SD | ↑ | P# | Macrodolichocephaly, facial alteration |
Weaver syndrome | ↑ | > 98th percentile | NA | NA | Abnormal facial alteration |
Simpson–Golabi–Behmel syndrome | ↑ | > 97th percentile | NA | NA | Macrocephaly, ocular hypertelorism (wide-spaced eyes) with broad upturned nose, macroglossia, and macrostomia (large mouth), supernumerary nipples, pectus excavatum, and hypotonia |
Perlman syndrome | NA° | 75th–97th percentile | NA° | NA° | Macrosomia, macrocephaly, round facies, hypotonia, visceromegaly, cryptorchidism and inguinal hernia |
Tatton–Brown–Rahman syndrome | NA | N/ > 2 SD | NA | P | Macrocephaly noticed at birth, joint hyperlaxity, scoliosis, hypotonia, and seizures |
Pituitary magnetic resonance imaging (MRI)
MRI is the gold standard for the evaluation of the pituitary gland in the paediatric and transition-age population [
146‐
148], providing morphological information and allowing the evaluation of size, signal characteristics, and vascularisation [
148]. The adenohypophysis is isointense to grey matter on non-contrast T1 and T2-weighted standard Spin Echo sequences [
149], whereas the neurohypophysis is characteristically hyperintense on T1 and hypointense on T2 sequences [
149]. An appropriate imaging protocol should include sagittal and coronal T1-weighted and T2-weighted sequences, as well as contrast-enhanced T1- weighted images following intravenous injection of gadolinium [
150]: normally, the pituitary gland enhances after gadolinium administration [
148,
149]. Due to the small dimensions of the sellar structures and potential intrinsic lesions, acquiring small field-of-view images is essential [
150], with either 2 or 3-mm sections obtained with 1.5 T scanning fields or 1.0 to 1.5-mm sections obtained with 3 T scanning[
150].
Pituitary tumours have typically delayed enhancement and washout characteristics [
148,
149]. Microadenomas are typically hypointense on both unenhanced and contrast-enhanced sequences, becoming iso/hyperintense to the normal pituitary gland in delayed sequences [
148]. Conversely, macroadenomas are usually isointense in T1-weighted images and present intense contrast enhancement after gadolinium injection [
148].
In adult patients with acromegaly, T2-hypointense adenomas are more common, smaller, and less invasive compared to T2-isointense and hyperintense tumours [
151]. Moreover, patients with T2-hypointense adenomas also have higher IGF1 values at baseline [
151,
152]. T2-weighted signal intensity is a marker for the granulation pattern [
153,
154]; accordingly, T2-hypointense adenomas have been linked with better hormonal responses and greater tumour shrinkage after presurgical somatostatin analogue administration [
151,
153,
155]. Currently, the few data reporting MRI findings in the paediatric and transition-age population mainly derive from retrospective studies. In the cohort of Rostmoyan et al
., the median age of rapid growth onset was 13 years (interquartile range 9–15), pituitary macroadenomas were more prevalent than microadenomas (84.3 vs 15.7%), with 15% of macroadenomas classified as ‘giant’ adenomas (> 4 cm); extrasellar extension was found in 89% of macroadenomas and extrasellar invasion in 64%. No differences were found between males and females [
39]. In another study, Colao et al
. reported data on the diagnosis and treatment of patients with GH-secreting adenomas with clinical onset in adolescence. Thirteen patients were enrolled, with a mean age of 17 ± 2 years; on MRI evaluation, the mean maximal tumour diameter was 21.8 ± 5.4 mm, and the mean tumour volume was 2756 ± 1895 mm
3 [
3,
41].
In conclusion, a pituitary MRI with an appropriate imaging protocol is mandatory for the evaluation of the pituitary gland in the paediatric and transition-age population with suspected gigantism (Fig.
2).
Conclusions
Human growth is a complex process and discriminating healthy tall children from those affected by acro-gigantism due to underlying diseases, either related to genetic (FIPA, and MEN1, more rarely CCS, MAS, 3PAs, MAX-associated tumours, NF1, and the recent PAM variants), or endocrine alterations (hyperthyroidism, obesity), is a compelling challenge. In general, females tend to receive a diagnosis of gigantism at a younger age than males, therefore males are more likely to be diagnosed during the transition age. A thorough clinical evaluation, using country- and disease-specific growth charts, is crucial before the biochemical assessment with GH and IGF1 measurements. Currently, a dedicated IGF1 reference range to guide the difficult differential diagnosis between constitutional tall stature and gigantism is still lacking. Nevertheless, the pathological effects of the prolonged exposition to supraphysiological levels of GH and IGF1 can cause systemic complications, mainly metabolic and cardiovascular; therefore, the clinical evaluation of pubertal staging and other signs and symptoms is of most importance, especially during the transition age.
Acknowledgements
This study has been proposed and scientifically supported by the TALENT Study Group, Sapienza University of Rome, Italy, and in particular: AM Savage, C. Foresta, C. Krausz, C. Durante, MC De Martino, D. Paoli, R. Ferrigno, S. Caiulo, M. Minnetti, V. Hasenmajer, C. Pozza, G. Kanakis, B. Cangiano, M. Tenuta, F. Carlomagno, A. Di Nisio, F. Pallotti, MG Tarsitano, M. Spaziani, F. Cargnelutti, I. Sabovic, G. Grani, C. Virili, A. Cozzolino, I. Stramazzo, T. Filardi. A. M. Savage, C. Foresta, C. Krausz, C. Durante, M. C. De Martino, D. Paoli, R. Ferrigno, S. Caiulo, M. Minnetti, V. Hasenmajer, C. Pozza, G. Kanakis, B. Cangiano, M. Tenuta, F. Carlomagno, A. Di Nisio, F. Pallotti, M. G. Tarsitano, M. Spaziani, F. Cargnelutti, I. Sabovic, G. Grani, C. Virili, A. Cozzolino, I. Stramazzo, T. Filardi.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.