# Clinical Predictors of Successful Pregnancy After In Vitro Fertilization (IVF): A Comprehensive Systematic Review of Evidence

## TL;DR

This paper reviews clinical factors that predict successful pregnancy after IVF, emphasizing the importance of female age, ovarian reserve, embryo quality, and sperm DNA integrity.

## Contribution

The study provides a systematic review of recent evidence on clinical predictors of IVF success, highlighting the role of machine learning models and the need for better validation.

## Key findings

- Female age is the most consistent predictor of IVF success, with a significant decline beyond 40 years.
- Embryo quality and sperm DNA fragmentation index are important predictors of successful pregnancy outcomes.
- Machine learning models show high accuracy in predicting IVF success but require further validation for clinical use.

## Abstract

In vitro fertilization (IVF) success is influenced by a complex interplay of patient- and treatment-related factors. Identifying reliable clinical predictors is crucial for patient counseling and individualized treatment planning. This systematic review aimed to synthesize the most recent evidence on clinical predictors of successful pregnancy after IVF. A comprehensive search of five electronic databases (PubMed, Scopus, Excerpta Medica database (Embase), Web of Science, and ClinicalTrials.gov) was conducted for studies published between 2020 and 2025. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, studies evaluating clinical predictors of clinical pregnancy or live birth in women undergoing IVF were included. Study selection, data extraction, and risk of bias assessment (using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool) were performed independently by two reviewers. A narrative synthesis was conducted due to significant methodological heterogeneity. Eight retrospective observational studies, comprising data from 40,490 IVF/intracytoplasmic sperm injection (ICSI) cycles, were included. Female age was the most consistent and powerful predictor, with a non-linear negative impact, particularly beyond 40 years. Ovarian reserve markers, anti-Müllerian hormone (AMH) and antral follicle count (AFC), were significant, with evidence suggesting AMH better predicts oocyte yield while AFC may better forecast embryo availability. Embryo quality parameters (number of high-quality embryos) were strongly associated with success. Male factors, including total progressive motile sperm count (TPMC) and sperm DNA fragmentation index (DFI), added incremental predictive value. Several studies developed predictive models using both traditional logistic regression and machine learning (ML) algorithms (e.g., eXtreme Gradient Boosting (XGBoost) and Random Forest), which demonstrated high accuracy but raised concerns regarding interpretability and temporal validity. Successful IVF pregnancy is multifactorial, with female age, ovarian reserve, embryo quality, and sperm DNA integrity being key prognostic determinants. While ML-based models show promise, their clinical integration requires rigorous external validation and transparency. Future research should prioritize prospective, multi-center designs and the integration of novel dynamic parameters to advance toward truly personalized prognostic tools in reproductive medicine.