胚胎形态动力学参数在胚胎整倍体预测模型构建中的应用研究
Application of Embryo Morphokinetic Parameters in the Construction of Embryonic Prediction Model
金真真 1魏泽锋 1于春娜 1牛学英 1王文华 1胡艳芬 1马亚兰 1王雪楠1
作者信息
- 1. 济宁医学院附属医院生殖医学科,山东 济宁 272029
- 折叠
摘要
目的:基于胚胎形态动力学参数建立整倍体预测模型,为辅助生殖技术中胚胎发育潜能的评估提供参考.方法:回顾性分析2020 年12 月至2023 年6 月接受胚胎植入前遗传学检测患者的临床资料,共纳入364 个周期的1383 个可利用囊胚.根据囊胚整倍性分为整倍体组(442 个囊胚)和非整倍体组(941 个囊胚).比较两组胚胎形态动力学参数,并进行多因素 logistic 回归分析,构建整倍体预测模型,采用ROC曲线分析整倍体预测模型的灵敏度和特异度.另选取2023 年 7 月至 2024 年 4 月纳入的215 个周期的658 个可利用囊胚的形态动力学参数资料进行临床验证,分析预测模型与临床实际的一致性.结果:整倍体组的t4、t5、s2、形成囊胚时间(tSB)、囊胚孵出时间(tHB)、tHB-tSB 均小于非整倍体组(P<0.05).logistic回归分析显示,t4、t5、s2、tSB、tHB、tHB-tSB 为囊胚整倍性的影响因素(P<0.05).整倍体预测模型:P=ex/(1+ex),X=-3.299-0.026×t4-0.018×t5-0.264×s2-0.009×tSB-0.007×tHB-0.153×tHB-tSB.Hosmer-Lemeshow拟合优度检验显示整倍体预测模型的拟合水平较好(χ2=8.700,P=0.368).整倍体预测模型的ROC曲线下面积为0.871(95%CI=0.852～0.889,P<0.001),灵敏度为75.86%,特异度为82.11%.临床验证整倍体预测模型的准确率为 79.94%,预测模型与临床实际的一致性中等(kappa=0.561).结论:胚胎形态动力学参数 t4、t5、s2、tSB、tHB、tHB-tSB 与整倍体形成具有相关性.本研究基于胚胎形态动力学参数构建的整倍体预测模型对胚胎整倍性具有较高的预测价值,可为胚胎发育潜能的评估提供参考.
Abstract
Objective:To establish a prediction model for euploidy based on embryo morphokinetic pa-rameters,so as to provide a reference for evaluating embryonic developmental potential in assisted reproductive technology.Methods:A retrospective analysis was performed on the clinical data of patients who underwent pre-implantation genetic testing between December 2020 and June 2023.A total of 1383 available blastocysts from 364 cycles were included.According to euploidy,the blastocysts were divided into the euploidy group(442 blastocysts)and the non-euploidy group(941 blastocysts).The morphokinetic parameters of the two groups were compared,and multivariate logistic regression analysis was performed.A prediction model for eu-ploidy was constructed.The ROC curve was used to analyze the sensitivity and specificity of the prediction model.Meanwhile,morphokinetic parameters of 658 available blastocysts from 215 cycles included from July 2023 to April 2024 were used for clinical validation.The consistency between the prediction model and clini-cal practice was analyzed.Results:t4,t5,s2,time of starting blastulation(tSB),time of hatching blastocyst(tHB),and tHB-tSB in the euploidy group were shorter than those in the non-euploidy group(P<0.05).Logistic regression analysis found that t4,t5,s2,tSB,tHB,and tHB-tSB were influencing factors of blasto-cyst euploidy(P<0.05).A prediction model for euploidy was constructed as follows:P==ex/(1+ex),X=-3.299-0.026×t4-0.018×t5-0.264×s2-0.009×tSB-0.007×tHB-0.153×tHB-tSB.Hosmer-Lemeshow goodness of fit test showed that the prediction model had good goodness of fit(χ2=8.700,P=0.368).The area under the ROC curve of the prediction model was 0.871(95%CI=0.852～0.889,P<0.001).Its sensi-tivity and specificity were 75.86%and 82.11%.Clinical validation indicated that the accuracy of the predic-tion model was 79.94%.The consistency between the prediction model and clinical practice was moderate(Kappa=0.561).Conclusion:Embryo morphokinetic parameters(t4,t5,s2,tSB,tHB and tHB-tSB)are related to the formation of euploidy.The prediction model constructed based on embryo morphokinetic parame-ters in this study is of high value in predicting embryo euploidy and can provide a reference for evaluating em-bryonic developmental potential.
关键词
整倍体/胚胎植入前遗传学检测/形态动力学/预测模型Key words
Euploidy/Pre-implantation genetic testing/Morphokinetics/Prediction model引用本文复制引用
基金项目
山东省自然科学基金青年项目(ZR2023QH478)
济宁市重点研发计划(2023YXNS099)
济宁市重点研发计划(2023YXNS228)
出版年
2024
国家科技期刊平台
NSTL
万方数据