Postmenarchal Adolescent Girls Demonstrate Multi-level Reproductive Axis Immaturity

Presentation Number: SUN-277
Date of Presentation: March 18, 2018, 2018

Bob Z. Sun, BA1, Tairmae Kangarloo, BS2, Judith Mary Adams, RN2, Patrick M. Sluss, PhD2, Donald Walt Chandler, PhD3, David T. Zava, PhD4, John A. McGrath, MA5, David M. Umbach, PhD1, Janet Elizabeth Hall, MD, MSc1, Natalie Shaw, MD1.
1National Institute of Environmental Health Sciences, Research Triangle Park, NC, USA, 2Massachusetts General Hospital, Boston, MA, USA, 3Endocrine Sciences, LabCorp, Calabasas Hills, CA, USA, 4ZRT Laboratory, Beaverton, OR, USA, 5Social and Scientific Systems, Inc., Durham, NC, USA.


Context: Menstrual irregularity is common after menarche. While it has been proposed that acquisition of estradiol (E2) positive feedback (+FB) is the last step in reproductive development, spontaneous and E2-induced LH surges of normal magnitude occur in premenarchal girls. We hypothesized that delayed maturation of other components of the reproductive axis explain menstrual irregularity.
Methods: Hormones (LH, FSH, E2, progesterone [P], and inhibin B [INHB]) were measured 3-5x/wk and follicle size assessed 2-4x/cycle in 23 girls aged 14.4 ± 0.2 y with gynecologic (gyn) age 0.6-3.5 y during 2 consecutive menstrual cycles. Hormone levels were centered to the mid-cycle surge of cycle 1 (MCS; day 0) and averaged at midcycle (-1 to +1), the luteal phase (LP; +2 to menses), and early follicular phase (EFP, wk1 of cycle 2) for analysis. E2 was also averaged during the late follicular phase (LFP; -3 to 0) to capture the E2 rise that starts 60h before the MCS. Data were compared with corresponding data from 65 adult historic controls (<35 y) with ovulatory cycles (OV) using repeated measures mixed models.
Results: Girls had anovulatory cycles (ANOV; 30%), OV with a short (<10d) luteal phase (short OV; 22%), or OV with normal luteal phase length (nl OV; 48%) with no group differences in cycle length, chronologic or gyn age, or BMI Z-score. All girls had lower LFP E2 than adults (ANOV 107.5 ± 18.4 [mean ± SE], p<0.01; short OV 140.4 ± 11.7, p=0.04, nl OV 146.6 ± 16.9 vs adult 214.5 ± 7.6 pg/ml; p<0.01), but nl OV girls still generated a normal LH surge (nl OV 20.0 ± 2.2 vs. adult 26.4 ± 1.1 IU/L, p=0.2; ANOV 10.6 ± 2.9, p<0.01; short OV 16.2 ± 2.0, p=0.046). In nl OV, the ratio of MCS LH to LFP E2 was similar to adults (p=0.3), consistent with mature E2 +FB, but the ratio was lower in ANOV (p<0.01) with a similar trend in short OV girls (p=0.09). Despite a normal surge and LP duration, nl OV did not attain adult LP levels of P (5.1 ± 0.8 vs 11.7 ± 0.4 ng/ml, p<0.01) or E2 (88.5 ± 13.2 vs 123.9 ± 4.2 pg/ml, p<0.01). FSH was lower in nl OV than in adults during the LP to EFP transition (LP 1.9 ± 0.3 vs 4.3 ± 0.2 IU/L, p<0.01; EFP 4.8 ± 0.5 vs 7.0 ± 0.2, p<0.01). Nl OV also had a slower dominant follicle growth rate than adults (1.5 ± 0.2 vs 2.3 ± 0.1 mm/d, p<0.01) and a smaller estimated pre-ovulatory follicle size (20.3 ± 0.8 vs 25.2 ± 0.3 mm, p<0.01). EFP INHB and mid- to LFP E2 were lower in nl OV girls vs. adults even after adjusting for FSH (INHB difference -84.7 ± 21.7 pg/ml, p<0.01) or FSH and follicle size (E2 difference -1.9 ± 1.2 pg/ml, p=0.01) respectively.
Conclusions: E2 +FB is impaired in girls with the most severe menstrual cycle abnormalities (ANOV, short OV). Despite robust E2 +FB, nl OV show diminished FSH secretion and ovarian responsiveness. Ovulation of a less mature follicle in nl OV is associated with decreased corpus luteum steroidogenesis. These results demonstrate that multiple components of the reproductive axis remain immature in the early post-menarchal years.


 B.Z. Sun: None. T. Kangarloo: None. J.M. Adams: None. P.M. Sluss: None. D.W. Chandler: Employee; Self; LabCorp. Stock Owner; Self; LabCorp. D.T. Zava: Owner/Co-Owner; Self; ZRT Laboratory. J.A. McGrath: None. D.M. Umbach: None. J.E. Hall: None. N. Shaw: None.