Enhancing predictions of embryo implantation in thin endometrium: a comparative analysis of endometrial thickness in frozen versus fresh IVF cycles

Adiel Kahana, Ittai Many, Yuval Fouks, Benny Almog, Yoav Baruch, Emily Hamilton, Foad Azem, Yoni Cohen*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

This study investigated whether the difference between endometrial thickness in the frozen embryo transfer (FET) cycle and the preceding fresh stimulated cycle predicts clinical pregnancy in women with thin endometrium. A retrospective analysis of 335 cycles in 227 women (age < 43) with endometrium < 8 mm in the FET cycle was conducted. Endometrial thickness was assessed using the ratio (ETfrozen/fresh) and absolute difference (ETfresh-frozen) between the maximal endometrial thickness measured during the stimulated IVF cycle and that measured in the FET cycle. We hypothesized that a smaller difference and higher ratio during FET will be associated with higher pregnancy rate. No embryo implantation occurred in women with an FET endometrial thickness < 5.9 mm and an ETfrozen/fresh < 0.86. Cycles with a positive beta-human chorionic gonadotropin (ß-hCG) showed a significantly lower mean ETfresh-frozen (1.89 ± 1.78 vs 2.30 ± 1.81, p = 0.03), and a significantly higher ETfrozen/fresh ratio (0.82 ± 0.17 vs. 0.77 ± 0.15, p = 0.024). Endometrial thickness in the FET cycle normalized to the preceding stimulated cycle is associated with implantation and may constitute a novel predictor of FET success in women with thin endometrium.

Original languageEnglish
Article number2404852
JournalHuman Fertility
Volume27
Issue number1
DOIs
StatePublished - 2024

Keywords

  • Thin endometrium
  • embryo-transfer
  • implantation
  • in vitro fertilization

Fingerprint

Dive into the research topics of 'Enhancing predictions of embryo implantation in thin endometrium: a comparative analysis of endometrial thickness in frozen versus fresh IVF cycles'. Together they form a unique fingerprint.

Cite this