Geng Tangyi, Ji Hui, Ding Kai, Yang Ye, Zhao Chun, Zhang Junqiang, Ling Xiufeng, Zhou Qiao
Department of Reproductive Medicine, The Affiliated Obstetrics and Gynaecology Hospital of Nanjing Medical University, Nanjing Women and Children's Healthcare Hospital, Nanjing, China.
Hum Fertil (Camb). 2025 Dec;28(1):2522886. doi: 10.1080/14647273.2025.2522886. Epub 2025 Jul 8.
The trend of delayed childbearing has increased the average age of parents, with the impact of paternal age on embryo euploidy remaining controversial. Therefore, this study aimed to investigate the impact of paternal age on embryo euploidy using retrospective clinical data and interpretable machine learning. This retrospective study included 960 couples and 4,718 embryos undergoing preimplantation genetic testing for aneuploidy (PGT-A). Couples were divided into two groups based on paternal age (Group 1 ≥ 40 years and Group 2 < 40 years). Statistical methods, including generalized estimating equation (GEE) and restricted cubic spline, were used to evaluate the relationship between paternal age and embryo euploidy. Interpretable machine learning models were employed to predict the likelihood of having at least one euploid embryo, validating the impact of paternal age on embryo euploidy. 867 and 3,851 blastocysts were selected as Group 1 and control Group 2, respectively. Couples with higher paternal age showed a significantly higher rate of embryo aneuploidy (60.21% vs. 41.03%, < 0.001). Logistic regression using GEE confirmed the association between paternal age and aneuploidy rate (OR: 1.396, 95% CI: 1.150-1.695, < 0.01). Combining clinical data analysis and interpretable machine learning models, the study provides evidence that paternal age negatively impacts embryo euploidy, emphasizing the need to consider paternal age in reproductive planning.
晚育趋势提高了父母的平均年龄，而父亲年龄对胚胎整倍体的影响仍存在争议。因此，本研究旨在利用回顾性临床数据和可解释的机器学习方法，探讨父亲年龄对胚胎整倍体的影响。这项回顾性研究纳入了960对夫妇和4718个接受非整倍体植入前基因检测（PGT-A）的胚胎。根据父亲年龄将夫妇分为两组（第1组≥40岁，第2组<40岁）。采用广义估计方程（GEE）和受限立方样条等统计方法，评估父亲年龄与胚胎整倍体之间的关系。使用可解释的机器学习模型预测至少有一个整倍体胚胎的可能性，以验证父亲年龄对胚胎整倍体的影响。分别选择867个和3851个囊胚作为第1组和对照组第2组。父亲年龄较大的夫妇胚胎非整倍体率显著更高（60.21%对41.03%，<0.001）。使用GEE的逻辑回归证实了父亲年龄与非整倍体率之间的关联（OR：1.396，95%CI：1.150-1.695，<0.01）。结合临床数据分析和可解释的机器学习模型，该研究提供了证据表明父亲年龄对胚胎整倍体有负面影响，强调在生殖规划中需要考虑父亲年龄。
