Sex Difference in Paediatric Growth Hormone Deficiency: Fact or Fiction?

Abstract

Short stature, defined as a height of less than the 2.3rd percentile or 2 standard deviations below the mean for age and sex, is a common indication for referral to paediatric endocrinology clinics [1]. In addition to the social implications of height, short stature with poor growth may be an indication of underlying systemic illness [2].

Though the estimated prevalence of growth hormone deficiency (GHD) is approximately 1:4000–10,000, the ultimate aim of a short stature evaluation in children who are growing below their genetic potential is to exclude this rare entity [3]. An integral part of patient evaluation, growth hormone provocative tests are widely used in clinical practice worldwide during short stature workup to diagnose GHD [4].

The medical literature is replete with publications that address the male predominance of growth hormone use both in non-GHD and GHD cases [5]. Various biases which contribute to this historically published male predominance are well documented. These include societal factors, since short stature in males is perceived as more concerning than in females; familial factors, as parents are often more concerned about the height of their sons than that of their daughters; and provider factors, as more boys than girls are referred for concerns about short stature [6, 7]. Given these biases, more boys than girls receive growth hormone provocative testing and are ultimately diagnosed with GHD [8].

The aforementioned biases undoubtedly prevent accurate assessment of the relative frequencies of paediatric GHD in males and females. If these biases are minimized, it is possible that the true frequencies of paediatric GHD are equal in males and females. This is also taking into account that although there may be certain rare pathologic conditions such as X-linked hypopituitarism, which could result in males with GHD when this and similar conditions are taken collectively, there still should not be a significant sex difference in GHD cases [9]. A closer look into sub-classifications of GHD utilized in clinical practice may offer an opportunity for evaluation of the true sex frequencies. These sub-classification frameworks may include (i) GHD severity based on peak growth hormone levels, (ii) the presence or absence of abnormal pituitary gland magnetic resonance imaging findings traditionally associated with a diagnosis of severe GHD such as the hypoplastic pituitary gland and (iii) the presence of GHD in the context of multiple pituitary hormonal deficiencies [10]. It should also be noted that although peak growth hormone levels are used as a surrogate for GHD severity, there is no association between this and with abnormalities such as the ectopic pituitary gland on pituitary magnetic resonance imaging [11]. Under another sub-classification framework, GHD is considered secondary to organic aetiology when it is associated with pituitary imaging abnormalities or multiple pituitary hormone deficiencies. Otherwise, it is idiopathic. Viewing GHD under this framework and considering that organic GHD does not seem to occur more frequently in males suggests that there may be an overdiagnosis of idiopathic GHD. This overdiagnosis may contribute to a perpetuation of the historically reported male predominance of GHD. This notion comes from the fact that when organic GHD is considered, this male predominance is lost [12].

Growth hormone secretion follows an ultradian rhythm, thus a single random sample does not accurately represent the state of the growth hormone-IGF I axis. To overcome this shortcoming of a random growth hormone measurement, growth hormone provocative testing is considered by many to be an integral part of diagnosing GHD. The insulin tolerance test, once viewed as the gold standard of growth hormone provocative testing, has now become obsolete due to safety concerns [13, 14]. Practices for growth hormone provocative testing differ widely between institutions in terms of both the specific agents utilized to elicit a peak growth hormone response and the growth hormone level below which GHD is diagnosed. Currently, a single cutoff of 7 ng/mL is used in many countries, whereas in the United States and Poland, a cutoff of 10 ng/mL is the norm [15]. Additionally, there are limitations that may affect peak growth hormone values derived from provocative testing in the paediatric population. Body mass index can impact the peak growth hormone values. Despite this, it is not usually considered. In addition, there is a lack of consensus regarding sex steroid priming, which can decrease the chance of a false positive GHD test result in a patient whose short stature is due to constitutional delay of growth and puberty. Since constitutional delay of growth and development and pubertal delay are more common in boys, this is of importance as patients with these diagnoses could potentially be misclassified as having GHD [16] Whether there needs to be different cutoffs for GHD diagnosis when sex steroid priming is used, and whether current provocative tests can fully discriminate short normal children from those with GHD, are unanswered questions that can impact the diagnostic accuracy of tests [17, 18]. Considering these limitations, coupled with the lack of correlation between GHD severity and peak growth hormone values, there may again be an overdiagnosis of idiopathic GHD with the current methodology.

Moreover, as the cutoff for the peak stimulated growth hormone value below which GHD is diagnosed has increased over the years, so has the number of patients diagnosed with GHD. This secular trend was initially fuelled by the advent of recombinant human growth hormone in the mid 1980s [19]. The diagnosis of GHD in the post-recombinant DNA-derived human growth hormone era became more common in contrast to the pre-recombinant DNA-derived human growth hormone era, during which the supply of growth hormone was outpaced by the demand, as at that time, growth hormone was cadaveric in origin [20]. This paradigm shift surrounding the diagnosis could perpetuate and may also contribute to an overdiagnosis of idiopathic GHD. As a testament to these two different time periods, during the pre-recombinant era, there was a height cap of 5 feet (150 cm) imposed in the United States, which ensured that only the shortest of individuals (arguably those with severe GHD) would be treated, which was different from the post-recombinant era in which the abundance of synthetic growth hormone has resulted in the creation of a multi-billion dollar industry. With this, it appears that the practice of medicine, disease diagnostics and prescribing practices have been influenced by drug availability [21, 22]

The highlighted concerns and weaknesses regarding provocative tests are likely important factors that have contributed to the historically published GHD sex differences. Also, the male predominance of a GHD diagnosis and treatment can be viewed within the lens of an ethical framework that accounts for the current practices regarding case definition, which lead to inequities surrounding access to care. This is particularly concerning for medically underserved girls (that theoretically would not supersede the societal impact in driving the male sex referral for growth evaluations), but nonetheless, who receive less evaluation for short stature and, as such, are likely to be diagnosed later with pathology [6]. Hence, the need for justice, as it pertains to equal access to growth hormone testing and GHD treatment for all individuals [20].

Moving the needle on the viewpoint of the differential perception regarding stature in boys when compared to girls from a societal and family standpoint, may be difficult to virtually impossible. However, from a provider standpoint, it is imperative to break free of the societal pressure to practice cosmetic endocrinology in favour of boys being taller. In addition, the decision to perform provocative testing should be primarily guided by auxology, especially considering that GHD should always be a clinical diagnosis. To achieve these goals, a change in the current practices from both the referring provider and the endocrinologist is needed.

It is important to get at the heart of the true sex frequencies of GHD through multi-center studies utilizing different diagnostic criteria for GHD. In addition, prospective studies and medical education campaigns targeting the elimination of the referral bias will prove essential in future studies. These will also minimize the biases that impact and contribute to the different sex frequencies seen in paediatric GHD. Ultimately, those efforts may promise the potential of improved health equity in the way we allocate resources and manage children who are referred and then evaluated for short stature.

The authors declare no conflicts of interest.