Frozen Embryo Transfer (FET) Cycles Treatment Outcome: Natural versus Artificial Preparation of Endometrium

Aim:
To assess cycle outcome of natural vs. artificial FET cycles.

Introduction:
Controlled ovarian stimulation may decrease endometrial receptivity due to supraphysiological E2 level in fresh IVF cycle, which may affect the treatment outcome. The most common preparation of the endometrium is a combination of estrogen supplementation with or without HPG axis downregulation. Conflicting results were reported regarding the various preparations.

Materials & Methods:
A retrospective cohort study of 711 FET cycles (500 patients) conducted between the years 2010-2017.Patients who underwent FET cycles were stratified into 3 groups: down regulation with GnRH -agonist and estrogen supplemented (GnRH Agonist-EE), Etinyl Estradiol (EE)-FET and natural cycle (NC)-FET. The demographic characteristics among the patients were compared. Treatment outcomes including clinical pregnancy rate (CPR) and live birth rate (LBR) were compared according to the study groups.

Results:
When comparing GnRH Agonist-EE-FET, EE-FET and NC-FET significant differences were found in peak estradiol levels prior to embryo transfer (355.39±451.93, 495.48±654.20, 214.30±209.54pg/ml, respectively; p<0.001), peak progesterone levels prior to transfer (0.43±0.28, 0.44±0.34, 0.71±0.54ng/ml, respectively; p<0.001). CPR was significantly higher in the NC-FET compared to EE-FET and GnRH Agonist-EE (46.1%, 32.9%, 23.3%, respectively; p=0.011) as well as LBR (35.2%, 25%, 15.3%, respectively; p=<0.001). Using multiple logistic regression controlling for confounders including patient age, peak estradiol levels, peak progesterone levels, endometrium preparation protocol, the NC -FET protocol was independently associated with higher CPR (OR =3.0, 95% CI 1.941-4.378; p<0.001) and LBR (OR=3.0, 95% CI 1.970-4.761; p<0.001).

Conclusion:
NC- FET preparation should be the first choice of treatment for FET cycles.