宫腔内人工授精早期妊娠丢失的Nomogram预测模型建立及其影响因素分析  

Establishment of Nomogram prediction model for early pregnancy loss after intrauterine insemination and analysis of its key factors

作　　者：关艺青 邓泽桦 赵晋 欧美珍[1] 何茜冬[1] 叶云 GUAN Yi-qing;DENG Ze-hua;ZHAO Jin;OU Mei-zhen;HE Qian-dong;YE Yun(The Assisted Reproductive Center,Zhongshan City People’s Hospital,Zhongshan528403)

机构地区：[1]中山市人民医院生殖中心,中山528403

出　　处：《生殖医学杂志》2023年第4期555-563,共9页Journal of Reproductive Medicine

摘　　要：目的基于临床数据分析宫腔内人工授精(IUI)后早期妊娠丢失的影响因素,根据影响因素建立Nomogram预测模型,用于评估IUI后早期妊娠丢失的风险。方法回顾性分析2014年1月至2022年7月于本院生殖中心行夫精IUI治疗的不孕症患者的临床资料,根据妊娠情况分为早期妊娠丢失组(51个周期)和继续妊娠组(170个周期),比较两组患者之间的基础资料。将70%患者随机分配至建模组,其余30%被分配到验证组,使用Logistic回归分析IUI后早期妊娠丢失的影响因素,并根据影响因素建立Nomogram预测模型;绘制ROC曲线,评估预测模型对IUI后早期妊娠丢失的的预测效能。结果早期妊娠丢失组和继续妊娠组在FSH/LH比值和精液处理方法的差异有统计学意义(P<0.05)。建模组单因素分析显示,对于早期妊娠丢失具有统计学意义的预测因素是精液处理方法、FSH/LH比值和男方乙肝病毒携带状态(P<0.05)。建模组多因素Logistic分析显示,具有统计学意义的预测因素是精液处理方法、FSH/LH比值、不孕类型和男方乙肝病毒携带状态(P<0.05)。在建模组中,预测模型的ROC曲线下面积(AUC)为0.701[95%CI(0.600,0.801)]。验证组显示出良好的区分度,AUC为0.658[95%CI(0.484,0.833)]。结论本研究建立了IUI后早期妊娠丢失的Nomogram预测模型;通过该预测模型,可以个性化地评估IUI早期妊娠丢失的风险。Objective:To analyze the influencing factors of early pregnancy loss after intrauterine artificial insemination(IUI)based on clinical data,and establish Nomogram prediction model according to the influencing factors to evaluate the risk of early pregnancy loss after IUI.Methods:The clinical data of the patients,with intrauterine insemination in Assisted Reproductive Center of Zhongshan People’s Hospital from January 2014 to July 2022 were retrospectively analyzed.According to the pregnancy status,the patients were divided into early pregnancy loss group(51 cycles)and continuing pregnancy group(170 cycles).The basic data of patients were compared between the two groups.70%of patients were randomly assigned to the model group and the remaining 30%were assigned to the validation group.Logistic regression was used to analyze the influencing factors of early pregnancy loss after IUI,and Nomogram prediction model was established according to the influencing factors.The ROC curve was drawn to evaluate the predictive effect of the Nomogram prediction model on early pregnancy loss after IUI.Results:There were significant differences in FSH/LH ratio and semen processing method between early pregnancy loss group and continuing pregnancy groups(P<0.05).Univariate analysis of the model group showed that the significant factors were the semen processing method,FSH/LH ratio and the paternal carrying status of hepatitis B virus(P<0.05).While multivariate analysis showed that the significant factors were the semen processing method,FSH/LH ratio,infertility type and the paternal carrying status of hepatitis B virus(P<0.05).In the model group,the area under the ROC curve(AUC)of the prediction model was 0.701[95%CI(0.600,0.801)].The validation group showed good discrimination with AUC of 0.658[95%CI(0.484,0.833)].Conclusions:The Nomogram prediction model for early pregnancy loss after IUI is established.The risk of early pregnancy loss after IUI can be evaluated specifically by the Nomogram prediction model.

关 键 词：宫腔内人工授精 早期妊娠丢失 流产 乙肝 Nomogram预测模型 

分 类 号：R711.6[医药卫生—妇产科学]