# Serum progesterone and vaginal gel dose in artificial-cycle frozen embryo transfer: association with pregnancy outcome

**Authors:** Hsiao-Chin Huang, Yu-Li Chuang, Chueh-Ko Yang, Horng-Der Tsai, Cheng-Hsuan Wu, Yu-Jing Chen, Hsin-Hung Wu

PMC · DOI: 10.3389/fendo.2026.1717071 · 2026-03-05

## TL;DR

This study found that higher doses of vaginal progesterone gel improved pregnancy outcomes in frozen embryo transfers, while blood progesterone levels on the day of transfer were not predictive.

## Contribution

The study demonstrates that vaginal progesterone gel dose, not serum levels, significantly affects pregnancy outcomes in artificial-cycle frozen embryo transfers.

## Key findings

- Serum progesterone levels on the day of FET did not predict clinical pregnancy or live birth rates.
- A higher Crinone dose (180 mg) was significantly associated with improved clinical pregnancy and live birth rates.
- Adjusting for baseline differences confirmed the benefit of higher progesterone gel doses.

## Abstract

To evaluate the clinical relevance of serum progesterone levels on the day of frozen embryo transfer (FET) and the dose of vaginal progesterone gel (Crinone) used for early luteal phase support (LPS) in artificial-cycle frozen embryo transfer (AC-FET).

This single-center retrospective study analyzed 342 AC-FET cycles in patients aged 18–45 at Changhua Christian Hospital, Taiwan (April 2018–December 2022). Women aged 18–45 years undergoing AC-FET were included; cycles with >3 prior failed FETs, major comorbidities, donor oocytes, or missing outcome data were excluded. Pregnancy outcomes were assessed per cycle. Crinone (90 mg or 180 mg daily) was administered for 5 days after endometrial thickness ≥7 mm and progesterone <1.5 ng/mL. Serum progesterone was measured on the day of FET. Clinical pregnancy rates based on gestational sac (CPR-S), fetal heartbeat (CPR-H), and live birth rate (LBR) were evaluated. Receiver operating characteristic (ROC) evaluated the predictive value of serum progesterone levels. Associations between study variables and pregnancy outcomes were assessed using univariate and multivariable logistic regression. Comparisons of pregnancy outcomes between the high- and low-dose Crinone groups were further conducted using inverse probability of treatment weighting (IPTW) to adjust for baseline differences between dosage groups.

Serum progesterone levels demonstrated poor predictive performance for CPR-S, CPR-H, and LBR (AUC range: 0.541–0.559) and were not independently associated with pregnancy outcomes in multivariable analyses. After adjustment using inverse probability of treatment weighting, a higher Crinone dose (180 mg vs. 90 mg) was significantly associated with higher CPR-S (OR, 1.87; 95% CI, 1.11–3.17; p = 0.020), CPR-H (OR, 2.11; 95% CI, 1.24–3.59; p = 0.006), and LBR (OR, 2.10; 95% CI, 1.31–3.38; p = 0.002).

Serum progesterone levels measured on the day of FET did not predict pregnancy outcomes in AC-FET cycles using vaginal progesterone gel. A higher Crinone dose was associated with higher early clinical pregnancy rates compared with a lower dose.

## Linked entities

- **Chemicals:** progesterone (PubChem CID 5994), Crinone (PubChem CID 5994)

## Full-text entities

- **Chemicals:** progesterone (MESH:D011374), Crinone (MESH:C400424)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Figures

2 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12999430/full.md

---
Source: https://tomesphere.com/paper/PMC12999430