Levonorgestrel emergency contraception : effects on endometrial development and embryo implantation

Abstract

Background: The dialog between the developing embryo and receptive endometrium is under the control of ovarian steroids and numerous biomolecules, some of which have been suggested as markers of endometrial receptivity. Unintended pregnancies are common. The standard treatment with levonorgestrel (LNG, 1.5mg either in a single dose or in two does with 12 hours intervals) provides women with a safe means of preventing unwanted pregnancy after unprotected intercourse. However, its mechanisms of action when used for emergency contraception (EC) remain a matter of discussion.

Overall aim: To study the effects of LNG used for EC on markers of endometrial receptivity, embryo implantation and first-trimester tissues.

Papers I and II: Endometrial biopsies were taken from fertil women with a regular menstrual cycle (n=12 for paper I; n=22 for paper II) during cycle days LH+4 to LH+5. The stromal and epithelial cells were isolated and cocultured in 3-dimensional endometrial constructs. Immunostaining of estrogen receptor (ER)-α and β, progesterone receptor (PR)- (A+B), vascular endothelial growth factor (VEGF), leukemia inhibitory factor (LIF), interleukin (IL)-1β, and cyclooxygenase (COX)-2 were present in both cultured epithelial and stromals cells, whereas the expression of PR-B, androgen receptor (AR), integrin αvβ3, and mucin 1 were confined to epithelial cells. Treatment with LNG did not change the expression of any markers studied or impair blastocyst attachment to the endometrial construct. Mifepristone, used as a positive control, up-regulated the ER-β and PR-B expression, whereas it down-regulated the expression of stromal VEGF, epithelial integrin αvβ3 and mucin 1. Mifepristone also inhibited blastocyst attachment in vitro.

Paper III: First-trimester decidua and chorionic villi were collected from women (n=9) who had self-administered LNG (1.5mg) for EC after ovulation and subsequently became pregnant and who opted to interrupt the pregnancy with vacuum aspiration. Samples from comparable, unexposed women (n=9) were collected as controls. No significant differences in the expression of ERs, PRs, AR or Ki67 in the samples exposed and unexposed to LNG were seen using immunohistochemistry.

Paper IV: Endometrial biopsies were taken from fertil women (n=8) on cycle days LH+6 to LH+8: i) throughout the normal cycle and ii) after treatment with LNG (0.75mg daily during LH+1 to LH+4). The treatment significantly reduced PR-A and PR-B immunoreactivity in glandular epithelium, whereas it increased LIF stromal immunoreactivity and mRNA expression. Differences in the expression of other markers of endometrial receptivity were insignificant between the two groups.

Conclusions: Levonorgestrel used for EC does not affect markers of endometrial receptivity studied neither in vitro nor in vivo. Levonorgestrel does not impair embryo attachment to a 3-D endometrial cell culture construct. Post-ovulatory administration of LNG has no effect on the expression of ovarian steroid recpetors or Ki67 in first-trimester tissues. New agents with improved effects on follicular development, ovulation and endometrial receptivity should be developed and available to increase the EC efficacy.