ESPE Abstracts

Background: Anti-Müllerian hormone (AMH) is produced by granulosa cells in small follicles prior to gonadotropin dependent growth and serum levels reflect the number of small antral follicles. There are currently no longitudinal data of individual AMH levels from infancy to adolescence.

Aim: To evaluate whether AMH in infancy and childhood is associated with AMH levels and ovarian morphology peripubertal and in adolescence, and whether high AMH in childhood is associated with a PCOS-like profile of reproductive hormones and ovarian morphology in adolescence.

Methods: A total of 695/1210 girls from the longitudinal Copenhagen Mother-Child Cohort were included in this nested cohort. Inclusion criteria: available AMH assessment; n=1483 AMH samples (mean samples pr. girl, n=3). Age at infancy: median (IQR) 0.26 (0.24-0.28), childhood: 7.4 (6.6-8.1), peripubertal: 11.4 (10.8-11.9) and adolescence: 16.1 (15.7-16.6) yrs. We assessed the individual dispersion of the AMH standard deviation score (SDS) from infancy to adolescence. Number of ovarian follicles was assessed by transabdominal ultrasound (TAUS) (Peripubertal and adolescence) and Magnetic Resonance Imaging (MRI) (Peripubertal).

Results: Each girl maintained her relative AMH SDS from infancy to adolescence (individual AMH SDS, mean 0.56±0.31 SD). Serum levels of AMH in infancy, childhood and peripubertal correlated with AMH in adolescence (infancy: n=115, Spearman r=0.347, P<0.001; childhood: n=145, r=0.637, P<0.001; peripubertal: n=204, r=0.675, P<0.001). Serum levels of AMH in infancy and childhood correlated with the number of small antral follicles (<7 mm) peripubertal and in adolescence; AMH infancy vs. TAUS peripubertal: n=77, r=0.253, P=0.026 and vs. MRI peripubertal: n=74, r=0.229, P=0.05; AMH childhood vs. TAUS peripubertal: n=78, r=0.461, P<0.001 and vs. MRI peripubertal: n=137, r=0.368, P<0.001; AMH childhood vs. TAUS adolescence: n=127, r=0.344, P<0.001. Girls with high AMH in childhood (AMH>30 pmol/l = highest 16% vs. other girls) had elevated levels of LH (median 4.53 vs. 3.29 IU/l, P=0.041) (Mann Whitney U-test), LH/FSH ratio (1.0 vs. 0.67, P=0.019), AMH (38.7 vs. 22.4 pmol/l, P<0.001), inhibin B (87 vs. 71 pg/ml, P=0.030), testosterone (1.05 vs. 0.81 nmol/l, P=0.005), and number of small follicles (2-7mm) (22 vs. 18, P=0.013) in adolescence.

Conclusions: Relatively stable individual levels of AMH from infancy to adolescence cause remarkable correlations of AMH in infancy and childhood with AMH levels as well as the number of AMH producing follicles peripubertal and in adolescence. These findings suggest an individual setpoint of ovarian activity of small antral follicles independently of gonadotropin stimulation throughout childhood. Furthermore, high AMH in childhood is associated with a PCOS-like profile in adolescence.