# Analysis of factors influencing clinical pregnancy rates in frozen-thawed embryo transfer cycles

## TL;DR

This study identifies key factors that influence the success of frozen-thawed embryo transfers, which could help improve pregnancy outcomes in assisted reproductive treatments.

## Contribution

The study introduces a novel combination of random forest and logistic regression to identify independent predictors of clinical pregnancy in frozen-thawed embryo transfer cycles.

## Key findings

- Clinical pregnancy rates were significantly higher for blastocyst transfers compared to cleavage-stage embryo transfers.
- Seven variables were identified as independent predictors of clinical pregnancy outcomes, including female age, number of high-quality blastocysts, and AMH levels.

## Abstract

To identify the determinants influencing clinical pregnancy outcomes in frozen-thawed embryo transfer (FET) cycles.

A retrospective analysis was conducted on patients who underwent FET at the Department of Reproductive Genetics, The First Affiliated Hospital of Kunming Medical University, between January 2018 and December 2023. A total of 7,302 FET cycles were included and categorized into two groups based on clinical pregnancy outcomes: the clinical pregnancy group (n = 3,365) and the non-clinical pregnancy group (n = 3,937). Baseline characteristics were compared between groups. A random forest algorithm was applied to rank the importance of variables, followed by dimensionality reduction using a sliding window sequential forward selection (SWSFS) method. The top-ranked predictors with the lowest average out-of-bag (OOB) error rates were incorporated into a multivariate logistic regression model to determine independent predictors of clinical pregnancy in FET cycles.

The overall clinical pregnancy rate (CPR) was 46.08%. The CPR was significantly higher in blastocyst transfers (61.14%) compared to cleavage-stage embryo transfers (34.13%) (χ² = 528.973, P < 0.01). The random forest model identified seven variables with the highest predictive value: female age, number of high-quality blastocysts, anti-Müllerian hormone (AMH) level, embryo stage at transfer, endometrial thickness on the day of transfer, number of high-quality cleavage-stage embryos, and pre-transfer endometrial thickness. Multivariate logistic regression analysis demonstrated that younger female age (OR: 0.93; 95% CI: 0.92–0.94), greater number of high-quality blastocysts (OR: 1.67; 95% CI: 1.49–1.88), higher AMH levels (OR: 1.03; 95% CI: 1.01–1.05), blastocyst transfer (OR: 2.31; 95% CI: 1.85–2.88), increased endometrial thickness on transfer day (OR: 1.10; 95% CI: 1.05–1.15), more high-quality cleavage-stage embryos (OR: 1.74; 95% CI: 1.59–1.92), and greater pre-transfer endometrial thickness (OR: 1.04; 95% CI: 1.00–1.09) were all independently associated with higher clinical pregnancy rates.

Female age, number of high-quality blastocysts, AMH levels, embryo stage at transfer, endometrial thickness on the day of transfer, number of high-quality cleavage-stage embryos, and pre-transfer endometrial thickness are significant predictors of clinical pregnancy outcomes in FET cycles. These findings may guide individualized embryo transfer strategies to optimize reproductive success.