The dose-dependent effect of estrogen on bone mineral density in trans girls

OBJECTIVE: Treatment in transgender girls can consist of puberty suppression (PS) with a gonadotropin-releasing hormone agonist (GnRHa) followed by gender-affirming hormonal treatment (GAHT) with estrogen. Bone mineral density (BMD) Z-scores decrease during PS and remain relatively low during GAHT, possibly due to insufficient estradiol dosage. Some adolescents receive high-dose estradiol or ethinyl estradiol (EE) to limit growth allowing comparison of BMD outcomes with different dosages. DESIGN: Retrospective study. METHODS: Adolescents treated with GnRHa for ≥1 year prior to GAHT followed by treatment with a regular estradiol dose (gradually increased to 2 mg), 6 mg estradiol or 100-200 µg EE were included to evaluate height-adjusted BMD Z-scores (HAZ scores) on DXA. RESULTS: Eighty-seven adolescents were included. During 2.3 ± 0.7 years PS, lumbar spine HAZ scores decreased by 0.69 [95% confidence interval (CI) -0.82 to -0.56)]. During 2 years HT, lumbar spine HAZ scores hardly increased in the regular group (0.14, 95% CI -0.01 to 0.28, n = 59) vs 0.42 (95% CI 0.13 to 0.72) in the 6 mg group (n = 13), and 0.68 (95% CI 0.20 to 1.15) in the EE group (n = 15). Compared with the regular group, the increase with EE treatment was higher (0.54, 95% CI 0.05 to 1.04). After 2 years HT, HAZ scores approached baseline levels at start of PS in individuals treated with 6 mg or EE (difference in 6 mg group -0.20, 95% CI -0.50 to 0.09; in EE 0.17, 95% CI -0.16 to 0.50) but not in the regular group (-0.64, 95% CI -0.79 to -0.49). CONCLUSION: Higher estrogen dosage is associated with a greater increase in lumbar spine BMD Z-scores. Increasing dosage up to 2 mg estradiol is insufficient to optimize BMD and approximately 4 mg may be required for adequate serum concentrations.