1 Introduction

Turner syndrome (TS) is a multisystem condition in phenotypically female individuals that results from absence of all or part of the second sex chromosome (Gravholt et al. 2017). The resulting haploinsufficiency for genes in the pseudo-autosomal region of the sex chromosomes, containing a key growth regulator, the SHOX (short stature homeobox-containing) gene is one of the factors causing disturbances of linear growth intrinsic to the phenotype of individuals presenting with TS (Binder 2011). Growth failure may be subtle in infancy and early childhood, but untreated growth failure is progressive, resulting in significant short stature in > 95% of affected girls and women. We describe the pattern of growth failure in TS, review safety and efficacy of growth hormone therapy in TS and summarize the recommendations in the updated international consensus guidelines for TS (Gravholt et al. 2024).

2 Growth in Turner Syndrome

Average height in non-growth hormone-treated adult women with TS ranges from approximately 138 cm to around 147 cm (−2.5 to −4.2 SD), generally regarded as ~20 cm below the mean height for reference female populations without TS (Gravholt et al. 2024; Lyon et al. 1985). Country-specific reference standards have been compiled for growth in TS (Isojima and Yokoya 2022) and have been included in the latest iteration of the guidelines (Gravholt et al. 2024). Recombinant human growth hormone (GH) has been the mainstay of treatment for short stature in TS since regulatory approval in the 1990s and has facilitated attainment of adult height close to the lower bounds of the reference female population (−2 SD) for many individuals who receive timely treatment. The purpose of GH treatment is to address growth failure and maintain a relatively normal height compared with peers during childhood, facilitating initiation of estrogen replacement at an age-appropriate time, and minimizing disability/stigmatization that may be associated with short stature in childhood, adolescence and adulthood. Attainment of an “average” height (relative to population norms) is not the intended goal of treatment and is not a realistic expectation for most.

Growth restriction in TS starts early in utero, leading to average birth weights and lengths that are below the normative means for gestational age and country of birth (Bernasconi et al. 1994; Davenport et al. 2002; Even et al. 2000; FitzSimmons et al. 1994). Linear growth further declines in the first few years of life, with short stature (defined as height below −2.0 SD for age and sex) established by 3–4 years of age on average (Davenport et al. 2002). Continued subnormal height gain during childhood, and the absence of a sex-steroid mediated pubertal growth spurt in the majority of individuals, leads to average adult heights in untreated individuals that are almost 3 SD below mean for the reference population (Isojima and Yokoya 2022). TS is associated with skeletal manifestations such as genu valgum, cubitus valgus, short fourth and fifth metacarpals/metatarsals, Madelung deformity (Binder et al. 2001) and disproportionate short stature, with relatively shorter limbs than trunk, resulting in increased sitting height-to-height ratio or upper-to-lower segment ratio (Rongen-Westerlaken et al. 1993). This disproportion is primarily attributed to haploinsufficiency of the SHOX gene, due to loss of the pseudo-autosomal region on the short arm of the second sex chromosome, on which the SHOX gene resides (Xp22.33) (Binder 2011; Clement-Jones et al. 2000; Rao et al. 1997). The exact mechanism of the skeletal dysplasia remains to be elucidated, but haploinsufficiency of the SHOX gene is thought to impact activation of SHOX-dependent target pathways involved in the regulation of chondrocyte differentiation and proliferation at the growth plate (Marchini et al. 2004, 2007; Munns et al. 2001, 2004). Altered physiology of the GH-dependent protein, insulin-like growth factor-I (IGF-I) (Lebl et al. 2001), reduced sensitivity to GH and IGF-I (Attie et al. 1997; Foster et al. 1994; Hochberg et al. 1997), as well as estrogen deficiency (Gravholt et al. 2001; Leung et al. 2004) may be additional contributory factors reducing linear growth in TS.

3 Effectiveness of Growth Hormone Treatment on Linear Growth

GH is the only currently approved therapy for short stature in TS. Based on two randomized, controlled trials to adult height that included either a placebo or a non-GH-treated group (Ross et al. 2011; Stephure 2005), GH treatment initiated after 5 years of age resulted in average height gain of approximately 5–7 cm more than the average height in the comparator groups. However, other studies in which treatment was started at younger ages or higher GH doses were utilized, have reported greater height gains (Carel et al. 1998; Sas et al. 1999). Average near-adult height gain of 1.07 SD (~7 cm) versus baseline was also reported in a long-term observational study of over 1600 participants with TS (Maghnie et al. 2022). A 2007 Cochrane review showed that in the first 1.0–1.5 years of treatment, GH-treated girls grew an average of 3 cm/year more than untreated girls (Baxter et al. 2007). Despite the brisk initial catch-up growth, height velocity (i.e., rate of linear growth) tends to decline over time, as it does in all GH-treated conditions, but remains above the untreated growth rate (Plotnick et al. 1998; Wasniewska et al. 2004).

The outcomes of GH therapy show significant individual variability, which may reflect biological variation (e.g., genetic height potential, as reflected by parental heights), combined with variation in treatment-related parameters (e.g., age at start of GH treatment, GH dose, adherence to treatment, age at onset or induction of puberty) (Ranke et al. 2007; Reiter et al. 2001). Nevertheless, individuals in the randomized control trials to adult height showed 40%–50% of GH-treated girls achieved heights within the normal adult range, compared with only 4%–16% of non-GH-treated controls (Ross et al. 2011; Stephure 2005). Younger age at GH initiation, taller baseline and parental heights, and longer duration (especially prepubertal) of GH therapy are associated with taller adult height outcomes (Ranke et al. 2007; Reiter et al. 2001; van Pareren et al. 2003). The karyotype (Ranke et al. 2000) and parent of origin of the X-chromosome (Devernay et al. 2012) were not reported to be predictive of GH response. One study showed that monosomy X and isochromosome Xq karyotypes were associated with lesser height gain in the second and third year of GH therapy compared with mosaic karyotypes (containing 46,XY, 46,XX cell lines) and deletion of Xq, despite height SD continuing to increase in all karyotype groups (Kasprzyk et al. 2021). This is in line with the observation that the increase in dosage of Xp (specifically the SHOX gene) would be associated with taller height outcomes. Other genetic variations have been proposed as markers of GH responsiveness in TS (Braz et al. 2014; Stevens et al. 2021) but have limited clinical utility at present.

In summary, the average height gains are around 1 cm for every year of GH treatment. It is important to recognize and share with families that height gain with GH therapy in TS is slow and modest over the duration of treatment. GH should be continued until epiphyseal fusion, as catch-down growth (decline across percentiles) is likely if treatment is discontinued early.

4 Quality of Life: Impacts of Height and Height Gain

Intrinsic to the practice of GH treatment of non-GH-deficient short children, including those with TS, is the concept that taller stature confers benefits to quality of life (QoL). But the analysis of QoL in general, and QoL in TS in particular, is complex. A large number of QoL studies have been conducted in TS, but attempts to compare and synthesize the results are hampered by many factors, including differences in study design, potential participation biases, control populations used, psychosocial instruments selected and outcomes evaluated, ages of subjects and timing of assessment, effect of comorbidities and concomitant therapies, country in which study was conducted and statistical methods employed. Critical to this latter point, no clinical trials have been reported in which QoL has been included as a primary outcome measure with adequate statistical power, and few have performed appropriate adjustments for the multiple comparisons undertaken in these studies.

QoL in TS appears to worsen with increasing age, as evidenced by a longitudinal Swedish study in which subjects were followed over 15–20 years. The investigators reported that older age in TS was associated with lower health related QoL on many of the subscales; however, there was no decline in health related QoL with age in the subset (~48%) who had previously received GH treatment (Krantz et al. 2019). Similar association between increasing age and worsening QoL were detected in other studies, and after adjusting for age, GH supplementation was associated with less pain but had no other impact on QoL (Amundson et al. 2010). Absolute height, and height gains attributed to GH treatment, have either neutral or somewhat positive associations with aspects of QoL (Bannink et al. 2006; Ertl et al. 2018; Jeż et al. 2018; Nadeem and Roche 2014; Naess et al. 2010; Rovet and Van Vliet 2019). The most striking results are those revealed by the only study in which QoL outcomes for GH-treated subjects were compared with those of a parallel non-GH-treated control group in the landmark Canadian trial (Rovet and Van Vliet 2019). Although overall QoL was similar for the 2 treatment groups, the GH-treated group showed strong correlations between adult stature and social competence, engagement and relations, time with friends, higher self-esteem, reduced teasing and less anxiety/depressed behavior; furthermore, greater height gain was associated with greater popularity and social relations. For the control group, whose adult height was on average ~ 7 cm shorter than that of treated subjects, height correlated only with school function and total self-concept. But the results also suggested that the positive impact on QoL may diminish over time as evidenced by a subset of this cohort who demonstrated no benefits of GH supplementation on QoL at age 20 years (Taback and Van Vliet 2011). A French study showed that QoL was not specifically associated with attained adult height or height gain but was impaired in those with cardiac disorders and hearing defects (Carel et al. 2005).

In summary, taller stature is associated with better QoL outcomes in some studies, but not others, and no detrimental effect of GH treatment has been detected. Height is only one of the factors that impact QoL in TS and it is important to set realistic expectations regarding the impact of GH supplementation on height gain and its impact on health related QoL in this population.

5 Timing of Growth Hormone Initiation

In the previous iteration of the international consensus guidelines for TS (Gravholt et al. 2017), treatment initiation was recommended around 4–6 years of age. Newer data from the Toddler Turner Extension Study (Quigley et al. 2021) showed that children who were treated early (by 2–4 years of age) were taller in childhood and at puberty compared to those who initiated treatment 1–2 years later. However, due to earlier discontinuation of GH with resulting catch down growth of some early-treated individuals, the total duration of GH therapy and hence near-adult heights were similar between groups. The authors concluded that early treatment initiation (before age 6 years) appears to counter the height decline that otherwise occurs from around 2 years of age, and thereby may help prevent childhood short stature and facilitate optimal timing of pubertal induction, by avoiding the need to prolong the prepubertal phase of growth (Quigley et al. 2021). The updated 2024 guidelines therefore recommend that GH may be offered as early as age 2 years, especially if there is evidence of growth failure (Gravholt et al. 2024), but timing of initiation should be individualized.

For individuals whose diagnosis of TS is delayed until pubertal age (11–13 years), GH may be initiated as long as epiphyses are open, and may be continued concomitant with low dose estrogen therapy for pubertal induction (Gravholt et al. 2024). GH treatment should be discontinued when there is little remaining growth potential, as evidenced by bone age ≥ 14 years and/or height velocity < 2 cm/year (Gravholt et al. 2024).

6 Growth Hormone Dosing and Adjustments

The usual starting dose of GH is 45–50 mcg/kg/day (~0.32–0.35 mg/kg/week) (Gravholt et al. 2024); higher initial doses are not routinely recommended, but if there is evidence of suboptimal response, dosage may be increased up to 68mcg/kg/day (~0.48 mg/kg/week) (Sas et al. 1999) in limited instances, based on tolerability, response and local regulatory guidelines. Poor adherence to treatment (Coutant et al. 2021), early discontinuation of treatment (Hughes et al. 2016) or lower prescribed GH doses (Cleemann Wang et al. 2020), result in reduced height gain and shorter adult height. The updated consensus guidelines recommend that treatment response should be monitored at least every 6 months, plotting the heights on both the standard population growth curve and TS-specific growth curves (Gravholt et al. 2024).

The guidelines endorse monitoring IGF-I at least annually, and suggest considering a reduction in the GH dose for persistently high IGF-I, substantially above the reference range for age, sex and pubertal stage (Gravholt et al. 2024). The long-term impact of elevated IGF-I and any potential association with neoplasia is unknown in children without underlying risk factors for neoplasia. On the other hand, potential IGF-I resistance in TS, and methodological challenges to precise and accurate measurements of IGF-I, accompanied by intra-individual and inter-laboratory variability, make the definition of a goal range for IGF-I challenging. Hence the recommendations support individualization of treatment goals and attention paid to consistency of IGF-I measurements, with some degree of tolerance for mild IGF-I elevations (Gravholt et al. 2024).

7 Safety of Growth Hormone Therapy in TS

GH therapy is generally well tolerated and safe in TS (Backeljauw et al. 2021, 2023; Maghnie et al. 2022). Known side effects of GH listed in product labeling, such as intracranial hypertension and slipped capital femoral epiphyses were reported more commonly during GH treatment in girls with TS (0.2% for each condition) compared with children treated for idiopathic short stature or idiopathic GH deficiency (0.1%) in one observational study (Bell et al. 2010). Overall prevalence of these events and serious adverse events related to GH is low (Backeljauw et al. 2021; Bell et al. 2010). Further, rates of these conditions in TS are similar to rates in children treated with GH for other growth disorders, such as organic forms of GH deficiency, or short stature associated with small for gestational age birth (Bell et al. 2010; Child et al. 2019).

Scoliosis is prevalent in individuals with TS (Kim et al. 2001; Marx et al. 2023) and more commonly seen in GH-treated individuals with TS, compared with other conditions for which GH therapy is prescribed (Bell et al. 2010). Worsening of preexisting scoliosis is associated with increase in linear growth, but GH therapy has not shown to be a risk factor for curve progression in one single institution retrospective study (Marx et al. 2023). Blood pressure, body composition and hearing loss were not adversely impacted by GH therapy in individuals with TS (Bannink et al. 2009; Davenport et al. 2010; Irzyniec et al. 2019; Mazzanti et al. 2005).

TS is associated with increased risk of both type 1 and type 2 diabetes (Sun et al. 2019) and some studies demonstrated a reversible increase in insulin resistance during GH therapy in TS, but without increase in prevalence of diabetes or impairment of glucose tolerance (Gravholt et al. 2002; Sas et al. 2000). Other studies have not shown a negative impact of GH therapy on insulin sensitivity or secretory capacity (Baronio et al. 2017) and the change in body composition with GH may be protective (Gnacińska et al. 2023; Wooten et al. 2008). Data from observational studies have also not shown an increase in diabetes related to GH therapy in TS (Bell et al. 2010; Maghnie et al. 2022). Nevertheless, given the increased risk of disorders of carbohydrate metabolism in girls and women with TS, irrespective of GH treatment, it is prudent to monitor glucose metabolism annually starting at age 10–12 years as recommended in the guidelines (Gravholt et al. 2024).

A higher prevalence of childhood lymphedema was noted in GH-treated vs. untreated women in one cross-sectional retrospective study, but the data do not imply causation, particularly as the GH-treated group had a greater prevalence of lymphedema at baseline (Irzyniec et al. 2019). There is only anecdotal evidence that GH may worsen pre-existing lymphedema in some individuals due to fluid retention. In the absence of conclusive evidence, therapeutic decisions on GH use need to be individualized in those who have pre-existing lymphedema and adjusted as needed if lymphedema worsens during treatment.

Evidence for any effect of GH on aortic diameter and risk of aortic dissection is inconclusive and this question warrants further study (Bondy et al. 2006; Duijnhouwer et al. 2019; Quezada et al. 2015; van den Berg et al. 2006).

No increase in cancer risk has been demonstrated in GH-treated individuals with TS (Backeljauw et al. 2023; Bolar et al. 2008). Data from an 8-country European cohort study comprising more than 24,000 patients treated with GH during childhood for a variety of growth disorders and followed for a total of over 400,000 patient-years, reported a 3-fold increase in mortality for patients with TS vs. general population mortality rates (Sävendahl et al. 2020). However, as pointed out by the authors, the elevated mortality rates could not be ascribed specifically to GH treatment, and comparison with the general population rather than untreated individuals with TS may have incorrectly estimated the mortality risk. Furthermore, it is well established that women with TS are at increased risk of mortality due to cardiovascular disease (Schoemaker et al. 2008). No increase in mortality has been reported from any of the observational registry studies in individuals with TS compared to other pediatric populations receiving GH therapy (Bell et al. 2010; Maghnie et al. 2022; Quigley et al. 2017).

8 Other Growth Promoting Agents

A few long-acting GH analogues are in clinical trials in various countries to evaluate safety and efficacy for treatment of short stature in TS. However, none of these products has received regulatory approval to-date, so presently only daily GH injections are available in most countries. Vosoritide, a C-type natriuretic peptide analog approved in the USA for treatment of short stature associated with achondroplasia, is hypothesized to improve growth in TS by enhancing SHOX-dependent signaling to promote chondrocyte differentiation and proliferation. One clinical trial of daily subcutaneous administration of vosoritide for 12 months is underway in the USA for demonstration of safety, efficacy and tolerability in TS (https://clinicaltrials.gov/).

The anabolic steroid, oxandrolone, has a modest synergistic effect with GH on adult height gain (~2–4 cm) (Mohamed et al. 2019; Sas et al. 2014; Sheanon and Backeljauw 2015) in TS, and was previously suggested for use as an off-label adjunctive therapy starting from age 10 years or older for girls whose response to GH alone was suboptimal (Gravholt et al. 2017). However, in 2023 the United States Food and Drug Administration withdrew approval for the marketing of oxandrolone in the USA for any indication, on the basis of unspecified “concerns over safety and effectiveness”. Although there was no indication that the concerns were related to the use of oxandrolone in TS, support for its use in the USA has been removed from the updated TS consensus guidelines (Gravholt et al. 2024).

Limb Lengthening procedures are not currently recommended in TS due to high complication rates for this procedure (Kim et al. 2014).

9 Estrogen and Pubertal Timing

The consensus guidelines (Gravholt et al. 2024) continue to encourage the initiation of estrogen replacement therapy in the form of 17β estradiol starting in low doses between the age of 11–12 years, in order to allow more timely development of secondary sexual characteristics, in individuals with confirmed primary ovarian insufficiency. The doses are progressively titrated up to full physiologic replacement over 2–4 years (Gravholt et al. 2024). As long as epiphyses remain open, concomitant treatment with GH may be pursued.

Pre-pubertal ultra-low dose estrogen has been studied in girls with TS as young as 5 years (Ross et al. 2011). This strategy, however is not recommended for growth promotion in the absence of a clear benefit and determination of optimal dose/formulation (Quigley et al. 2002). Low-dose estradiol (below feminizing dosages) is expected to initially stimulate growth velocity synergistically with GH therapy in TS (Hasegawa et al. 2017; Rosenfield et al. 1998; Ross et al. 2011). Doses of estradiol in the early-to-mid-pubertal range subsequently promote a modest pubertal growth spurt, and higher doses result in eventual fusion of epiphyses.

10 Conclusion

In summary, short stature is a prominent feature of TS, and daily GH injections are currently the mainstay of growth promotion therapy in this condition. Substantial data obtained over nearly 40 years confirm efficacy for increasing childhood growth and adult height, with no unexpected safety concerns. However, the impact of GH therapy on quality-of-life measures remains inconclusive. Hence decisions on GH therapy should be considered in the context of the child's height potential, and incorporate the family's values and goals, with shared decision making and an individualized treatment plan.

Author Contributions

Roopa Kanakatti Shankar: conceptualization, original draft, writing, reviewing and editing; Charmian A. Quigley: writing, reviewing and editing; Tsuyoshi Isojima: reviewing and editing; Nelly Mauras: reviewing and editing; Steven D. Chernausek: reviewing and editing; Malgorzata Wasniewska: reviewing and editing; Theo C. J. Sas: reviewing and editing.

Conflicts of Interest

Roopa Kanakatti Shankar has received research funding from BioMarin for an investigator-initiated clinical trial of vosoritide in Turner syndrome, paid to the institution, Children's National Hospital, Washington DC; will serve as a consultant for BioMarin, and is a member of the steering committee of the InsighTS registry, sponsored by the Turner Syndrome Global Alliance. Charmian A. Quigley was a former employee of Eli Lilly and Company, a manufacturer of recombinant GH. Nelly Mauras has institutional research grants from Novo Nordisk and has received funding support from Pfizer. Steven D Chernausek serves as a scientific advisor for Ascendis Pharma. Malgorzata Wasniewska is a consultant for Pfizer, Merck, Sandoz and Novo Nordisk. Charmian A. Quigley, Tsuyoshi Isojima, and Theo C. J. Sas have no conflicts of interest to declare.