Abstract
Context
Growth is an important topic for many transgender boys. However, few studies have investigated the impact of puberty suppression (PS) and gender-affirming hormone treatment (GAHT) on growth and adult height.
Objective
To evaluate the effect of PS and GAHT on growth and adult height.
Design
Retrospective cohort study.
Setting
Specialized gender identity clinic.
Participants
A total of 146 transgender boys treated with GnRH analogues and testosterone who reached adult height.
Main outcome measures
Growth, bone age (BA), adult height, and difference between adult height and predicted adult height (PAH) and midparental height.
Results
In those with BA ≤14 years at start (n = 61), a decrease in growth velocity and bone maturation during PS was followed by an increase during GAHT. Adult height was 172.0 ± 6.9 cm; height SD score was similar to baseline (0.1; 95% CI, −0.2 to 0.4). Adult height was 3.9 ± 6.0 cm above midparental height and 3.0 ± 3.6 cm above PAH at start of PS. A younger BA at start PS was associated with an adult height significantly further above PAH.
Conclusion
During PS, growth decelerated followed by an acceleration during GAHT. Although adult height SD score was similar to baseline, adult height was taller than predicted based on BA at baseline, especially in those who started treatment at a younger BA. It is reassuring that PS and GAHT do not have a negative impact on adult height in transgender boys and might even lead to a slightly taller adult height, especially in those who start at a younger age.
Keywords: growth, gender dysphoria, transgender, testosterone, adult height, puberty suppression, GnRHa
Gender dysphoria involves an incongruence between the experienced gender and the sex assigned at birth (1). Medical treatment for adolescents consists of puberty suppression (PS) and subsequent gender-affirming hormone treatment (GAHT). PS with GnRH analogues (GnRHa) inhibits the development of undesired secondary sex characteristics and gives adolescents more time to explore their gender identity. If the gender-dysphoric feelings persist during the puberty-suppressive phase, GAHT may be initiated from the age of approximately 16 years to induce development of secondary sex characteristics congruent with the affirmed gender (1).
In transgender boys (female sex assigned at birth, male gender identity), GAHT consists of intramuscular or transdermal testosterone. This induces physical changes such as increased facial and body hair, increased muscle mass, and lowering of the voice. Growth and predicted adult height are also topics of interest during consultation. Because Dutch males reach a mean adult height 13.1 cm taller than Dutch females (2), height is also considered a sex characteristic. When treatment is initiated before adult height has been reached, transgender boys may ask how treatment will impact their growth and whether they will reach an adult height within the male population range. However, although the impact of GnRHa and GAHT on growth in transgender girls has recently been evaluated (3), the effect on growth in transgender boys is still unknown.
The limited data available in transgender boys solely describe the effects of GnRHa on growth. Several studies have observed a decrease in height SD score (SDS) (4–6). Furthermore, it remains unclear whether testosterone can induce catchup growth and reverse the changes in height SDS that occur during GnRHa treatment.
The timing of treatment initiation might also play a role in the effects on growth. It has been suggested that treatment with PS and GAHT might have a negative impact on adult height (7). Therefore, some clinicians argue that postponing treatment with GnRHa until growth is (nearly) finished may result in a taller adult height. This poses a dilemma for young transgender boys between achieving maximal growth and preventing the development of unwanted secondary sex characteristics such as breast development. The latter may give great distress because of the need to wear binders for several years and the wish to undergo a mastectomy (8).
Another hypothesis is that the effect of testosterone on growth is dose dependent. Rapid increase of the testosterone dose may cause rapid acceleration of bone maturation, thereby shortening the period of growth. This may compromise adult height. In contrast, a more gradual increase of testosterone dosage may induce an increase in growth rate during a longer period, leading to an increased adult height, which is why a different dose schedule is proposed for adolescents that have and have not finished linear growth (1). However, the effects of different testosterone dose schedules on growth and adult height in transgender adolescents have not been studied.
With this cohort study, we aim to evaluate the effect of GnRHa and testosterone treatment on growth in transgender boys. Furthermore, the impact of timing of treatment, tempo of dose increase, and body mass index (BMI) will be investigated. We hypothesize that PS will result in a decrease of growth velocity and GAHT will result in subsequent catchup growth that may be greater when the testosterone dose is slowly increased. Additionally, we expect these effects to be larger in subjects with a younger bone age (BA) at the start of PS.
Materials and Methods
Subjects
This study is part of the Amsterdam Cohort of Gender dysphoria (ACOG) study which includes the complete population of all ages seen at the gender identity clinic of the Amsterdam University Medical Center, location VUmc from 1972 until December 2018 (9). Transgender boys were eligible for inclusion if they had started PS before age 16 years, received testosterone treatment for a minimum of 6 months, and if they had reached the age of 18 years at the time of data collection. Transgender boys were excluded if they had not reached adult height defined as height at skeletal age ≥14 years or a growth velocity <2 cm per year. The cohort was divided into 2 subgroups based on growth potential. The pubertal group consisted of subjects with BA ≤14 years at start of PS or, if BA was not available, menarche <1 year before start PS. The postpubertal group, which served as a control group, consisted of subjects who had BA >14 years or menarche ≥1 year before start of PS.
Treatment Protocol
Adolescents were diagnosed with gender dysphoria by mental health professionals according to the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fourth and later Fifth Edition (10, 11). All individuals received PS for at least 6 months followed by GAHT from the age of 15 to 16 years. PS consisted of GnRHa (triptorelin 3.75 mg every 4 weeks subcutaneously or IM; or 11.25 mg every 10 to 12 weeks intramuscularly). GAHT consisted of intramuscular testosterone injections with a starting dose of 25 mg/m2every 2 weeks. As described in the Endocrine Society Guidelines, the dosage was gradually increased every 6 months leading up to an adult dose of 125 mg every 2 weeks or 250 mg every 3 to 4 weeks (1). If transgender boys had (nearly) completed endogenous puberty before the start of treatment, the testosterone dose was increased more rapidly, starting with 75 mg every 2 weeks which was increased to 125 mg every 2 weeks after 6 months. Treatments were individualized by treating clinicians. Before 2014, subjects were treated with GnRHa until gonadectomy, which nearly all individuals chose to undergo because this was required before a legal sex change was possible. After a law change in 2014, it became more common to leave the ovaries in situ. Thereafter, the protocol was adapted and GnRHa was discontinued when an adult dose of testosterone was reached.
Measurements
Height and weight were assessed every 3 to 6 months from start of PS. Height was measured using a wall-mounted stadiometer, and weight was measured using a digital floor scale. Height SDS was calculated according to Dutch reference data from Schönbeck et al. (2) and BMI SDS was calculated according to reference data from Cole et al. (12). Midparental height was calculated using the following formula: Midparental height = (paternal height + maternal height)/2 – 6.5 (13). BA was determined at start of PS and at start of GAHT by evaluating X-rays of the left hand using female reference data from the Greulich and Pyle atlas (14). Predicted adult height (PAH) was determined using the female prediction tables according to Bayley and Pinneau (15). Parental height and BA were not routinely obtained in individuals in whom it was clinically obvious that they had attained their adult height.
Laboratory Investigations
To determine serum testosterone levels, a competitive immunoassay (Architect, Abbott, Abbott Park, IL, USA) was used (interassay coefficient of variation [CV] 6%-10%, lower limit of quantification [LOQ] 0.1 nmol/L) from January 2013. Serum estradiol levels were measured using liquid chromatography–tandem mass spectrometry (VUmc) with an interassay coefficient of variation of 7% and an LOQ of 20 pmol/L from July 2014. All values ≤20 pmol/L were entered as 20 pmol/L. Serum testosterone and estradiol levels from before these dates were converted to Architect and liquid chromatography–tandem mass spectrometry values respectively as described by Wiepjes et al (16). From April 2012, the chemiluminescence immunoassay (LIAISON, DiaSorin) was used to determine IGF-1 (interassay coefficient of variation 7%). IGF-1 measurements before April 2012 were converted to Liaison values as described by Boogers et al (3). LH and FSH were measured using an immunometric assay (Delfia, PerkinElmer) until June 2011 (interassay CV <7%, LOQ U/L for LH and interassay CV <5%, LOQ U/L for FSH). After June 2011, an immunometric assay (Architect, Abbott) was used (interassay CV <6%, LOQ 0.1 U/L for LH and interassay CV <5%, LOQ 0.1 U/L for FSH) using the formula Architect = 0.91*Delfia−0.01 for conversion of LH and Architect = 1.04 Delfia + 0.001 for conversion of FSH. Blood withdrawals were untimed in relation to the administration of medication because they were taken at the day of the appointment with the clinician.
Statistical Analyses
All data were analyzed using STATA 15.1 (StataCorp, College Station, TX, USA). Categorical data are presented as number (%) and continuous variables as mean ± SD for normally distributed values, or median (interquartile range [IQR]) for nonnormally distributed data.
Change in height and height SDS between start of PS and start of GAHT were evaluated using mixed model regression analysis with measurements clustered within participants with height variables as outcome and time as determinant. The use of mixed model analyses allows for use of all available data points and has been proven to be a reliable method to handle missing data (17). Mixed-model analysis was used to analyze the change in height SDS during PS between subjects with young BA (≤12 years) and more mature BA (>12 years) at start PS. Height SDS at start GAHT was compared between both groups using linear regression analysis. Mixed-models analysis was used to analyze the change in BA minus chronological age (BA – CA) and in PAH between start PS and start GAHT.
Changes in height and height SDS between start of GAHT and adult height were analyzed using mixed-model regression. This also applied to the differences in height and height SDS between start PS and adult height. The overall effect of PS and GAHT on growth was evaluated by comparing adult height with midparental height and PAH at start PS using linear regression analysis. The difference between adult height and PAH at start PS was adjusted for BA at start PS and BMI SDS separately using linear regression analysis. This was also performed using multivariable regression.
The effect of the testosterone regimen on growth was evaluated by comparing subjects in whom the testosterone dose was increased until adult dose (125 mg/2 weeks) in less than 1 year vs more than 1 year. Linear regression analysis was used to compare the baseline characteristics and adult height of the subjects between the 2 groups. Linear regression was also used to assess the effect of time to adult dose as a continues variable on the difference between adult height and PAH at start of PS and GAHT.
Ethical Approval
The protocol for data collection for the ACOG dataset was assessed by the local medical ethical committee who determined that the Medical Research Involving Human Subjects Act did not apply to this data collection. The need for informed consent was waived because of the retrospective design of the study and the size of the cohort.
Results
The ACOG dataset consisted of 8831 individuals, 3481 of whom were assigned female at birth and of whom 1320 had their first appointment before the age of 18 years. After selection for GnRHa use (n = 693), testosterone use (n = 533), age ≥18 years at last visit from which data were available (397), GnRHa initiation before the age of 16 years (148), exclusion of 1 subject who had not reached adult height, and 1 subject who was treated with oxandrolone, a total of 146 transgender boys were included.
Baseline Characteristics
The study population (n = 146) was divided into 2 subgroups based on growth potential. The pubertal group consisted of 61 transgender boys with a mean age of 12.7 ± 1.0 years and BA of 12.4 ± 1.0 at start of PS. The postpubertal group with little or no growth potential consisted of 85 individuals with a mean age of 15.1 ± 0.9 years and BA of 15.7 ± 1.1 at start PS (Table 1). The outcomes of the pubertal group will be described in detail. Data from the postpubertal group will be briefly described at the end of the Results section.
Table 1.
Baseline characteristics of the pubertal group (those with bone age ≤14 years at start of PS) and of the postpubertal group (those with bone age >14 years at start of PS)
| Characteristic | Pubertal group (n = 61) | Postpubertal group (n = 85) | Total (n = 146) |
|---|---|---|---|
| Age at start PS (y) | 12.7 ± 1.0 | 15.1 ± 0.9 | 14.1 ± 1.5 |
| Height at start of PS (cm) | 158.3 ± 8.5 | 166.4 ± 6.7 | 162.9 ± 8.5 |
| ȃMissing | 0 (0) | 4 (5) | 4 (3) |
| BMI at start of PS | 18.8 ± 2.4 | 22.4 ± 3.8 | 20.4 ± 3.6 |
| ȃMissing | 0 (0) | 33 (39) | 33 (23) |
| BMI SDS at start of PS | 0.3 ± 1.1 | 0.9 ± 1.1 | 0.6 ± 1.1 |
| ȃMissing | 0 (0) | 33 (39) | 33 (23) |
| Bone age at start of PS (y) | 12.4 ± 1.0 | 15.7 ± 1.1 | 13.7 ± 1.9 |
| ȃMissing | 5 (8) | 48 (56) | 53 (36) |
| PAH at start of PS (cm) | 169.0 ± 7.6 | 167.3 ± 6.8 | 168.3 ± 7.3 |
| ȃMissing | 5 (8) | 48 (56) | 53 (36) |
| Midparental height (cm) | 168.1 ± 6.2 | 166.4 ± 5.7 | 167.2 ± 6.0 |
| ȃMissing | 5 (8) | 26 (31) | 31 (21) |
| Tanner breast stage at start of PS | |||
| ȃB2 | 8 (13) | 0 (0) | 8 (5) |
| ȃB3 | 20 (33) | 0 (0) | 20 (14) |
| ȃB4 | 23 (38) | 10 (12) | 33 (23) |
| ȃB5 | 8 (13) | 59 (69) | 67 (46) |
| ȃMissing | 2 (3) | 16 (19) | 18 (12) |
| Menarche before start of PS | |||
| ȃYes | 20 (33) | 81 (95) | 100 (68) |
| ȃNo | 39 (64) | 0 (0) | 40 (27) |
| ȃMissing | 2 (3) | 4 (5) | 6 (4) |
Data are presented as number (%) or mean ± SD.
Abbreviations: BMI, body mass index; PS, puberty suppression; PAH, predicted adult height; SDS, SD score.
Puberty Suppression
The mean duration of PS was 3.1 ± 0.9 years. At baseline, mean height was 158.3 ± 8.5 cm (female height SDS +0.1 ± 1.5, male height SDS −0.1 ± 1.5). During the puberty suppressive phase, height increased by 8.6 cm (95% CI, 7.5-9.6) to 166.9 ± 7.0 cm at start of GAHT. Female height SDS decreased to −0.2 ± 1.0 (decrease of −0.2; 95% CI, −0.5 to 0.1). Transgender boys with BA >12 years at start PS declined more in height SDS during PS compared with transgender boys with BA ≤12 years (difference between groups −0.6; 95% CI, −0.7 to −0.4) (Fig. 1), but height SDS at start of GAHT did not differ between the groups (difference 0.3; 95% CI, −0.3 to 0.9). Additionally, PS resulted in a deceleration of bone maturation. At the beginning of PS, BA was comparable to CA with a BA – CA of −0.3 ± 0.9 years. This difference was greater among those with BA ≤12 at start PS compared with those with BA >12 years (difference, 0.7 years; 95% CI, 0.2-1.2). At start of GAHT, the difference between BA and CA further increased to −2.2 ± 1.2 years (change since start of PS −1.9 years; 95% CI, −2.2 to −1.6). PAH increased by 2.4 cm (95% CI, 1.3-3.4) between start PS and start GAHT.
Figure 1.
Effect of puberty suppression on female height SDS in subjects with BA ≤12 years (n = 32, 57%) and BA >12 years (n = 24, 43%) at start of PS with 95% confidence intervals. PS, puberty suppression.
Hormone Treatment
GAHT was initiated at a mean age of 15.8 ± 0.5 years. From the start of GAHT, height increased by 5.0 cm (95% CI, 4.3-5.8) to an adult height of 172.0 ± 6.9 (female height SDS +0.2 ± 1.1, male height SDS −1.7 ± 1.0) (Fig. 2). Female height SDS increased by 0.3 SDS from the start of GAHT to adult height (95% CI, 0.2-0.5) (Table 2 and Fig. 3). However, adult height SDS did not differ from height SDS at the start of PS (difference, 0.1; 95% CI, −0.2 to 0.4). Compared with male reference data, 22 subjects (36%) had an adult height below −2.0 SDS. BA – CA remained stable in the first 18 months of GAHT but after that, BA progressively increased. In 45 subjects (74%), adult height was taller than PAH at the start of PS. Adult height 3.9 ± 6.0 cm above midparental height (95% CI, 2.4-5.4) and 3.0 ± 3.6 cm above PAH at start of PS (95% CI, 2.0-3.9), but adult height was close to PAH at the start of GAHT (difference, 0.2 ± 2.3 cm; 95% CI, −0.5 to 0.9).
Figure 2.
Mixed model analysis of height during PS and GAHT of 17 subjects who initiated PS at age 12 and GAHT at age 16, plotted on the growth chart for Dutch girls from Schönbeck et al (2). GAHT, gender-affirming hormone therapy; PS, puberty suppression.
Table 2.
Growth during puberty suppression and testosterone treatment in the pubertal and postpubertal group
| Pubertal group (n = 61) | Postpubertal group (n = 85) | Pubertal vs postpubertal group (95% CI) | |
|---|---|---|---|
| Height (cm) | |||
| ȃStart of PS | 158.3 ± 8.5 | 166.4 ± 6.7 | 8.0 (5.5-10.6) |
| ȃMissing | 0 (0) | 4 (5) | |
| Start of GAHT | 166.9 ± 7.0 | 167.5 ± 7.0 | 0.6 (−1.7 to 2.9) |
| ȃMissing | 0 (0) | 3 (4) | |
| ȃAdult height | 172.0 ± 6.9 | 169.0 ± 6.8 | 3.0 (0.7-5.2) |
| Female height SDS | |||
| ȃStart of PS | +0.1 ± 1.5 | −0.1 ± 1.0 | 0.1 (−0.3 to 0.6) |
| ȃMissing | 0 (0) | 4 (5) | |
| ȃStart of GAHT | −0.2 ± 1.0 | −0.2 ± 1.1 | 0.0 (−0.3 to 0.4) |
| ȃMissing | 0 (0) | 3 (4) | |
| ȃAdult height | +0.2 ± 1.1 | −0.3 ± 1.1 | 0.5 (0.1-0.8) |
| PAH at start of GAHT (cm) | 172.3 ± 7.5 | 168.9 ± 7.5 | 3.5 (0.0-7.0) |
| ȃMissing | 18 (30) | 53 (62) | |
| ȃΔ Adult height—PAH (cm) | |||
| ȃPAH at start PS | 3.0 ± 3.6 | 1.8 ± 2.0 | 1.2 (−0.1 to 2.4) |
| ȃMissing | 5 (8) | 48 (56) | |
| ȃPAH at start of GAHT | 0.2 ± 2.3 | 1.4 ± 1.6 | 1.2 (0.2-2.1) |
| ȃMissing | 18 (30) | 53 (62) | |
| ȃΔ Adult height—midparental height (cm) | 3.9 ± 6.0 | 3.0 ± 4.7 | 0.9 (−1.0 to 2.9) |
| ȃMissing | 5 (8) | 26 (31) | |
| ȃΔ BA – CA (cm) | |||
| ȃStart of PS | −0.3 ± 0.9 | 0.9 ± 1.0 | 1.2 (0.8-1.6) |
| ȃMissing | 5 (8) | 48 (56) | |
| ȃStart of GAHT | −2.2 ± 1.2 | 0.1 ± 1.0 | 2.4 (1.8-2.9) |
| ȃMissing | 18 (30) | 53 (62) |
Data are presented as number (%) or mean ± SD.
Abbreviations: BA, bone age; CA, chronological age; GAHT, gender-affirming hormone treatment; PAH, predicted adult height; PS, puberty suppression; SDS, SD score.
Figure 3.
Effect of hormone treatment on female height SDS in transgender boys in whom adult testosterone dose was reached within <1 year (n = 21) compared with transgender boys in whom adult dose was reached after ≥1 year (n = 40) with 95% confidence intervals. GAHT, gender-affirming hormone therapy.
Factors Influencing Growth
When comparing subjects who reached the adult testosterone dose within 1 year (n = 21, 10.8 months to adult dose [IQR, 9.5-11.6]) to those in whom the adult dose was reached after 1 year (n = 39, 13.6 months to adult dose [IQR, 12.8-15.9]), no differences in change in height SDS during GAHT were observed (Fig. 3). Additionally, adult height did not differ between the 2 dosing regimens (difference between groups 0.8 cm; 95% CI, −3.0 to 4.6). Both groups were comparable at baseline regarding age, height, and BA at the start of PS, midparental height, and PAH at start PS. When analyzed as a continuous variable, time to adult dose was not associated with the difference between adult height and PAH at the start of PS and PAH at the start of HT either (data not shown).
When looking at the difference between adult height and PAH at the start of PS, an inverse correlation with BA at start PS was found (Fig. 4). A younger BA at the start of PS was associated with an adult height further above PAH (1.2 cm/y; 95% CI, 0.3-2.1). A similar correlation was observed between BA at start GAHT and the difference between adult height and PAH (1.2 cm/y; 95% CI, 0.5-2.0). Those with a younger BA at the start of PS also had an adult height further above midparental height, but this correlation did not reach statistical significance (1.3 cm/y; 95% CI, −0.4 to 3.0).
Figure 4.
Diagram of the correlation between bone age at start puberty suppression and difference in adult height and PAH at start PS. Each circle represents the data for 1 subject. β , regression coefficient; PAH, predicted adult height; PS, puberty suppression.
A lower BMI SDS at the start of PS was associated with an adult height further above PAH at the start of PS. Per 1 SDS decrease in BMI, the adult height was 0.4 cm further above PAH at the start of PS, but this was not statistically significant (95% CI, 0.0-0.8). In a multivariable analysis, associations between BA and BMI SDS at start PS on the 1 hand and the difference between adult height and PAH at start PS on the other hand were similar to those in the univariable analysis.
Laboratory Investigations
During the puberty suppressive phase, gonadotrophins were suppressed (median LH 0.3 U/L [IQR, 0.3-0.4]; median FSH 1.1 U/L [IQR, 0.8-1.7]). Median serum estradiol levels were 20 pmol/L, which was the lower limit of detection (IQR, 20-20). IGF-1 levels did not significantly change after initiation of PS and remained stable during PS. Mean testosterone levels in the first year of GAHT were not significantly higher in subjects who reached adult dose within 12 months compared with >12 months (95% CI, −18.2 to 18.5 nmol/L). When adult testosterone dose was reached, median serum testosterone levels were 20 nmol/L (IQR, 11-35) and free testosterone was 599 pmol/L (IQR, 338-1179). IGF-1 levels slightly increased by 3.6 nmol/L (95% CI, 0.9-6.4) after initiation of testosterone and remained stable thereafter with a mean serum level of 32.3 ± 5.9 nmol/L. No significant differences in IGF levels during GAHT between participants who reached adult dose within 12 months and those who reached adult dose >12 months were observed.
Adolescents With Little or no Growth Potential (Postpubertal Group)
In total, 85 transgender boys had little or no growth potential at the start of PS (Table 1). The mean duration of PS was 1.2 ± 0.8 years. Height at start PS was missing in 4 subjects. From the start of PS, height increased by 2.5 cm (95% CI, 2.1-3.0) to an adult height of 169.0 ± 6.8 cm (female height SDS −0.3 ± 1.1, male height SDS −2.1 ± 1.0) (Table 2). In 70 (82%) subjects, height increased ≥1 cm during treatment. Adult height was 3.0 ± 4.7 cm above midparental height (n = 57) and 1.8 ± 2.0 cm above PAH at start PS (n = 37). When comparing adult height in the pubertal group to that in the postpubertal group, individuals in the pubertal group were taller (difference, 3.0 cm; 95% CI, 0.7-5.2). However, the difference between adult height and midparental height was not significantly different between the groups (difference, 0.9 cm; 95% CI, −1.0 to 3.0).
Discussion
This study is the first to investigate the effect of both GnRHa and testosterone treatment on growth and adult height in transgender boys. We observed a decrease in growth velocity and bone maturation during PS and an increase in growth during GAHT. Adult height was taller than predicted at the start of the treatment, and this difference was more pronounced when the treatment was initiated earlier. In contrast, the tempo of the testosterone dose increase did not affect growth.
During the puberty suppressive phase, we observed a mean decrease in height SDS of −0.2. Ghelani et al reported no significant changes in height SDS of GnRHa treatment (6). This was because their subjects were older (median age of 16.6 years) and were only followed for 1 year. Our data showed a greater decrease in height SDS among those with more advanced BA (>12 years) at the start of PS. In subjects with younger BA height, SDS did not decline. This may be due the fact that the group with BA >12 years consisted of more early maturers, who were relative tall for their age, with a regression to the mean after the initiation of GnRHa treatment. This is supported by the fact that baseline height SDS was higher in the group with BA >12 years but height SDS at the start of GAHT was similar in the groups with BA ≤12 and >12 years. Our results are similar to findings in a previous study by Weise et al in girls with central precocious puberty treated with GnRHa (18). A lower growth velocity in girls with BA ≥10 years was reported compared with girls with BA <10 years. Besides the retarded growth, bone maturation was delayed by −1.9 years during PS. This decreased bone maturation during GnRHa treatment has also been described in studies in girls with central precocious puberty (19, 20).
After the decrease in height SDS during PS, subsequent catchup growth during testosterone treatment was observed. Growth acceleration after initiation of testosterone is also described in a previous study by Stoffers et al who reported a height SDS increase from −0.1 ± 1.0 at start GAHT to 0.1 ± 0.8 SDS after 2 years of testosterone in transgender boys (21). The slightly smaller increase described by Stoffers et al might be explained by the fact that most subjects had completed linear growth before start of treatment.
The effect of GnRHa and testosterone on adult height in transgender boys was analyzed in 4 manners: (1) the change in height SDS during the treatment; (2) the difference between adult height and PAH at start PS; (3) the difference between adult height and midparental height; and (4) comparison of growth data from the pubertal group with the “control” group of postpubertal adolescents with little/no growth potential at start of the treatment.
Adult height SDS was comparable to height SDS at start PS, which would imply that PS and GAHT have no impact on adult height.
In contrast, adult height was 3.0 ± 3.6 cm above PAH at start PS. This deviation was greater than the slight underestimation of adult height by 1.2 ± 4.1 cm and 0.4 ± 4.3 cm described by Zachmann et al in healthy girls with an age of 12 and 13 years, respectively, using the same method of Bayley and Pinneau to predict adult height (22). This might suggest a slight enhancing effect of PS and GAHT on adult height. This is further supported by the finding that adult height was 3.9 ± 6.0 cm above midparental height.
Finally, the fact that subjects from the pubertal group were 3.0 cm taller than those with little/no growth potential at the start of treatment also suggests a positive effect on growth. However, when comparing the difference between midparental height and adult height in both groups, no significant differences were found. This might be explained by the fact that midparental height was missing in 31% of the postpubertal group. It is possible that physicians enquired about parental height more often when subjects were shorter to try and understand if short stature was familial.
When all 4 methods to evaluate the effect on adult height are combined, we conclude that treatment with PS and GAHT does not negatively influence adult height in transgender boys. When comparing adult height to height predicted at the start of treatment, to midparental height and to height in the control group, there even seems to be a slight positive effect of PS and GAHT on growth.
When looking at factors that might influence growth, tempo of testosterone dose increase was not found to affect growth or adult height. In boys with constitutional delay of growth and puberty and central hypogonadotropic hypogonadism it is thought that higher doses of testosterone induce accelerated skeletal maturation and thereby compromise adult height (23–25). We did not observe such an effect in transgender boys, nor was there a significant difference in testosterone levels, but this might also be attributed to the untimed blood tests resulting in a wide variation of serum testosterone levels. Based on these findings, gradually increasing the dose over the first year of GAHT seems to allow adolescents to attain their growth potential. However, we did not investigate side effects of testosterone treatment, like mood changes, aggression, and increased hematocrit, which may be more common with a fast dose increase (21). This should also be taken into account when deciding on an optimal schedule to induce puberty.
Interestingly, we found that a younger BA at start of PS had a positive impact on the difference between adult height and PAH at start PS. This might partly be due to the slight underestimation of adult height using the prediction method of Greulich and Pyle, which has been reported to be larger in younger girls (22). However, this does not account for the 1.2 cm increase of adult height per year that PS is started earlier that we describe. Furthermore, adult height was also slightly further above midparental height in those with a younger BA, which also indicates a positive impact of starting treatment at a younger BA on adult height. This contradicts findings in girls with early-normal puberty in whom treatment with GnRHa did not result in an increase of adult height (26). An explanation might be found in the different treatment regimens. Puberty suppression until the age of 16 years followed by puberty induction with testosterone may have different effects on the epiphyseal growth plates compared with a shorter period of PS followed by endogenous puberty with estradiol.
Although not statistically significant, there was a trend toward a negative association between BMI SDS at start PS and the difference between adult height and PAH at start PS. Earlier studies also described a negative effect of higher BMI on adult height by accelerated growth and earlier epiphyseal closure (27, 28). The absence of a statistically significant association in our study might be explained by the fact that BMI was within the normal range in most subjects.
Since 36% of the transgender boys had an adult height < −2 SDS compared with the male population, which many, in our clinical experience, are sad about, future research should investigate possible therapeutic options to enhance adult height. Because estradiol has an important role in epiphyseal closure, aromatase inhibition alongside GnRHa and testosterone might be an effective treatment (29). However, there are no efficacy or safety data on such an approach in transgender boys, and it may have a negative impact on bone mineral accrual, which is already attenuated by GnRHa treatment. An American study by Grimstad et al reported a 5 cm taller height in transgender boys treated with oxandrolone, an androgen which cannot be aromatized to estrogen, compared with untreated individuals (30). However, the relatively small groups in this study were not comparable regarding age at the start of treatment. The use of oxandrolone, which is currently not available in Europe, or aromatase inhibitors to promote adult height might be promising but needs further research.
This study has strengths and limitations. Strengths are the reasonable cohort size and the comparison with individuals who (nearly) reached adult height before the start of treatment as controls. The retrospective character of the study is a limitation because this resulted in some missing data. Other limitations are that X-rays for BA were assessed by various radiologists, possibly resulting in interobserver variability, and that hormone regimens were individualized, making it more difficult to compare treatment strategies.
Conclusion
This study has provided new information on the impact of GnRHa and testosterone treatment on growth and adult height, which can be used to counsel transgender boys. PS resulted in decelerated growth and bone maturation, which accelerated after the initiation of testosterone. Although the growth pattern was altered by PS and GAHT, the treatment does not have a negative impact on adult height. There may even be a slight positive effect on adult height that is more pronounced in those who start GnRHa at a younger age. Nonetheless, adult height was below −2 SDS compared with the general male population in 36%. Future research is needed to assess the efficacy and safety of possible treatment options for those who have a strong wish to increase adult height into the male reference range, such as oxandrolone or aromatase inhibitors.
Abbreviations
- ACOG
Amsterdam Cohort of Gender
- BA
bone age
- BMI
body mass index
- CA
chronological age
- CV
coefficient of variation
- GAHT
gender-affirming hormone treatment
- GnRHa
GnHR analogue
- IQR
interquartile range
- LOQ
limit of quantification
- PAH
predicted adult height
- PS
puberty suppression
- SDS
SD score
Contributor Information
Lieve Anne Willemsen, Center of Expertise on Gender Dysphoria, Amsterdam UMC location Vrije Universiteit Amsterdam, 1081 HV Amsterdam, the Netherlands; Department of Endocrinology, Amsterdam UMC location Vrije Universiteit Amsterdam, 1081 HV Amsterdam, the Netherlands; Department of Pediatric Endocrinology, Amsterdam UMC location University of Amsterdam, 1105 AZ Amsterdam, the Netherlands.
Lidewij Sophia Boogers, Center of Expertise on Gender Dysphoria, Amsterdam UMC location Vrije Universiteit Amsterdam, 1081 HV Amsterdam, the Netherlands; Department of Endocrinology, Amsterdam UMC location Vrije Universiteit Amsterdam, 1081 HV Amsterdam, the Netherlands; Amsterdam Gastroenterology Endocrinology Metabolism, 1105 BK Amsterdam, the Netherlands.
Chantal Maria Wiepjes, Center of Expertise on Gender Dysphoria, Amsterdam UMC location Vrije Universiteit Amsterdam, 1081 HV Amsterdam, the Netherlands; Department of Endocrinology, Amsterdam UMC location Vrije Universiteit Amsterdam, 1081 HV Amsterdam, the Netherlands; Amsterdam Gastroenterology Endocrinology Metabolism, 1105 BK Amsterdam, the Netherlands.
Daniel Tatting Klink, Division of Pediatric Endocrinology, Department of Pediatrics, Ghent University Hospital, Ghent, Oost-Vlaanderen 9000, Belgium.
Adrianus Sarinus Paulus van Trotsenburg, Department of Pediatric Endocrinology, Amsterdam UMC location University of Amsterdam, 1105 AZ Amsterdam, the Netherlands; Amsterdam Gastroenterology Endocrinology Metabolism, 1105 BK Amsterdam, the Netherlands.
Martin den Heijer, Center of Expertise on Gender Dysphoria, Amsterdam UMC location Vrije Universiteit Amsterdam, 1081 HV Amsterdam, the Netherlands; Department of Endocrinology, Amsterdam UMC location Vrije Universiteit Amsterdam, 1081 HV Amsterdam, the Netherlands; Amsterdam Gastroenterology Endocrinology Metabolism, 1105 BK Amsterdam, the Netherlands.
Sabine Elisabeth Hannema, Center of Expertise on Gender Dysphoria, Amsterdam UMC location Vrije Universiteit Amsterdam, 1081 HV Amsterdam, the Netherlands; Department of Pediatric Endocrinology, Amsterdam UMC location University of Amsterdam, 1105 AZ Amsterdam, the Netherlands; Amsterdam Gastroenterology Endocrinology Metabolism, 1105 BK Amsterdam, the Netherlands.
Data Availability
The dataset generated during and analyzed during the current study is not publicly available because of privacy regulations.
References
- 1. Hembree WC, Cohen-Kettenis PT, Gooren L, et al. Endocrine treatment of gender-dysphoric/gender-incongruent persons: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2017;102(11):3869–3903. [DOI] [PubMed] [Google Scholar]
- 2. Schönbeck Y TH, van Dommelen P, Bakker B, et al. The world's tallest nation has stopped growing taller: the height of Dutch children from 1955 to 2009. Pediatr Res. 2013;73(3):371–377. [DOI] [PubMed] [Google Scholar]
- 3. Boogers LS, Wiepjes CM, Klink DT, et al. Trans girls grow tall: adult height is unaffected by GnRH analogue and estradiol treatment. J Clin Endocrinol Metab. 2022;107(9):e3805–e3815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Schagen SE, Cohen-Kettenis PT, Delemarre-van de Waal HA, Hannema SE. Efficacy and safety of gonadotropin-releasing hormone agonist treatment to suppress puberty in gender dysphoric adolescents. J Sex Med. 2016;13(7):1125–1132. [DOI] [PubMed] [Google Scholar]
- 5. Schulmeister C, Millington K, Kaufman M, et al. Growth in transgender/gender-diverse youth in the first year of treatment with gonadotropin-releasing hormone agonists. J Adolesc Health. 2022;70(1):108–113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Ghelani R, Lim C, Brain C, Fewtrell M, Butler G. Sudden sex hormone withdrawal and the effects on body composition in late pubertal adolescents with gender dysphoria. J Pediatr Endocrinol Metab. 2020;33(1):107–112. [DOI] [PubMed] [Google Scholar]
- 7. Roberts SA, Carswell JM. Growth, growth potential, and influences on adult height in the transgender and gender-diverse population. Andrology. 2021;9(6):1679–1688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. van de Grift TC, van Gelder ZJ, Mullender MG, Steensma TD, de Vries AL, Bouman M-B. Timing of puberty suppression and surgical options for transgender youth. Pediatrics. 2020;146(5):e20193653. [DOI] [PubMed] [Google Scholar]
- 9. Wiepjes CM, Nota NM, de Blok CJM, et al. The Amsterdam cohort of gender dysphoria study (1972-2015): trends in prevalence, treatment, and regrets. J Sex Med. 2018;15(4):582–590. [DOI] [PubMed] [Google Scholar]
- 10. American Psychiatric Association . Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association Publishing; 2000. [Google Scholar]
- 11. American Psychiatric Association . Diagnostic and Statistical Manual of Mental Disorders. 5th ed. American Psychiatric Association Publishing; 2018. [Google Scholar]
- 12. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320(7244):1240–1243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Tanner JM, Goldstein H, Whitehouse RH. Standards for children's height at ages 2-9 years allowing for heights of parents. Arch Dis Child. 1970;45(244):755–762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Greulich WW, Pyle SI. Radiographic Atlas of Skeletal Development of the Hand and Wrist. 2nd ed. Stanford University Press; 1959. [Google Scholar]
- 15. Bayley N, Pinneau SR. Tables for predicting adult height from skeletal age: revised for use with the Greulich-Pyle hand standards. J Pediatr. 1952;40(4):423–441. [DOI] [PubMed] [Google Scholar]
- 16. Wiepjes CM, de Blok CJ, Staphorsius AS, et al. Fracture risk in trans women and trans men using long-term gender-affirming hormonal treatment: a nationwide cohort study. J Bone Miner Res. 2020;35(1):64–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Twisk J, de Boer M, de Vente W, Heymans M. Multiple imputation of missing values was not necessary before performing a longitudinal mixed-model analysis. J Clin Epidemiol. 2013;66(9):1022–1028. [DOI] [PubMed] [Google Scholar]
- 18. Weise M, Flor A, Barnes KM, Cutler GB Jr, Baron J. Determinants of growth during gonadotropin-releasing hormone analog therapy for precocious puberty. J Clin Endocrinol Metab. 2004;89(1):103–107. [DOI] [PubMed] [Google Scholar]
- 19. Clemons RD, Kappy MS, Stuart TE, Perelman AH, Hoekstra FT. Long-term effectiveness of depot gonadotropin-releasing hormone analogue in the treatment of children with central precocious puberty. Am J Dis Child. 1993;147(6):653–657. [DOI] [PubMed] [Google Scholar]
- 20. Boepple PA, Mansfield MJ, Crawford JD, Crigler JF Jr, Blizzard RM, Crowley WF Jr. Gonadotrophin-releasing hormone agonist treatment of central precocious puberty: an analysis of growth data in a developmental context. Acta Paediatr Scand Suppl. 1990;367:38–43. [DOI] [PubMed] [Google Scholar]
- 21. Stoffers IE, de Vries MC, Hannema SE. Physical changes, laboratory parameters, and bone mineral density during testosterone treatment in adolescents with gender dysphoria. J Sex Med. 2019;16(9):1459–1468. [DOI] [PubMed] [Google Scholar]
- 22. Zachmann M, Sobradillo B, Frank M, Frisch H, Prader A. Bayley-Pinneau, Roche-Wainer-Thissen, and Tanner height predictions in normal children and in patients with various pathologic conditions. J Pediatr. 1978;93(5):749–755. [DOI] [PubMed] [Google Scholar]
- 23. Rogol AD. New facets of androgen replacement therapy during childhood and adolescence. Expert Opin Pharmacother. 2005;6(8):1319–1336. [DOI] [PubMed] [Google Scholar]
- 24. Young J. Approach to the male patient with congenital hypogonadotropic hypogonadism. J Clin Endocrinol Metabol. 2012;97(3):707–718. [DOI] [PubMed] [Google Scholar]
- 25. Boehm U, Bouloux P-M, Dattani MT, et al. European Consensus statement on congenital hypogonadotropic hypogonadism—pathogenesis, diagnosis and treatment. Nat Rev Endocrinol. 2015;11(9):547–564. [DOI] [PubMed] [Google Scholar]
- 26. Bangalore Krishna K, Fuqua JS, et al. Use of gonadotropin-releasing hormone analogs in children: update by an international consortium. Horm Res Paediatr. 2019;91(6):357–372. [DOI] [PubMed] [Google Scholar]
- 27. Brener A, Bello R, Lebenthal Y, Yackobovitch-Gavan M, Phillip M, Shalitin S. The impact of adolescent obesity on adult height. Horm Res Paediatr. 2017;88(3-4):237–243. [DOI] [PubMed] [Google Scholar]
- 28. Sopher AB, Jean AM, Zwany SK, et al. Bone age advancement in prepubertal children with obesity and premature adrenarche: possible potentiating factors. Obesity (Silver Spring). 2011;19(6):1259–1264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Wit JM. Should skeletal maturation be manipulated for extra height gain? Front Endocrinol (Lausanne). 2021;12:812196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Grimstad FW, Knoll MM, Jacobson JD. Oxandrolone use in trans-masculine youth appears to increase adult height: preliminary evidence. LGBT Health. 2021;8(4):300–306. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The dataset generated during and analyzed during the current study is not publicly available because of privacy regulations.