Nomogram to predict live birth in infertile women with adenomyosis undergoing IVF/ICSI treatment: a retrospective cohort study

TL;DR

This study created a tool to predict live birth chances for women with adenomyosis undergoing IVF/ICSI treatments.

Contribution

A well-calibrated nomogram was developed specifically for infertile women with adenomyosis undergoing IVF/ICSI.

Findings

The predictive model achieved an AUC of 0.915 in the modeling cohort and 0.940 in the validation cohort.

Variables like age, BMI, and endometrial thickness were key predictors of live birth outcomes.

The model showed excellent calibration between predicted and actual outcomes.

Abstract

The study aimed to develop a nomogram to predict live birth outcomes in infertile women with adenomyosis undergoing IVF/ICSI treatment.

Data were collected from the Center for Reproductive Medicine, Shandong University, between January 2016 and December 2020. This study included 222 fresh embryo transfer (ET) cycles and 266 frozen embryo transfer (FET) cycles. Patients were divided into a modeling cohort and a validation cohort. In the modeling cohort, multivariable logistic regression was used to generate the predictive model and construct the nomogram. The predictive model was subsequently validated and calibrated.

In total, 249 patients (51.0%) achieved clinical pregnancy, and 165 patients (33.8%) had a live birth. Predictive variables in the final model included mean initial uterine diameter, age, body mass index (BMI), type of infertility (primary or secondary), stage of transferred embryo, endometrial thickness, number of embryos transferred, and pregnancy type. The area under the receiver operating characteristic (ROC) curve for the final prediction model was 0.915 (95% confidence interval [CI]: 88.6%–94.3%) in the modeling cohort and 0.940 (95% CI: 90.3%–97.6%) in the validation cohort. Calibration curves and the Hosmer–Lemeshow (H–L) test demonstrated excellent consistency between predicted and actual live birth outcomes.

This study established a well-calibrated predictive model capable of accurately forecasting live birth outcomes in infertile women with adenomyosis undergoing IVF/ICSI treatment.

Figures