摘要
目的 探讨前置胎盘伴胎盘植入的高危因素并构建列线图预测模型.方法 回顾性选取2020年5月至2023年1月入同济大学附属妇产科医院产检并剖宫产分娩的前置胎盘妊娠孕妇306例作为研究对象.根据是否伴有胎盘植入分为单纯前置胎盘组和前置胎盘伴胎盘植入组.收集研究对象的一般资料(年龄、体重指数、孕次、人工流产次数、剖宫产次数、原手术医院等级)、既往妊娠情况(剖宫产分娩史、宫腔操作史、保胎史)、孕期临床表现(产前有无阴道流血)、B超检查结果(前置胎盘类型、胎盘厚度).采用Logistic回归分析患者发生前置胎盘伴胎盘植入的高危因素,根据高危因素构建列线图预测模型,并采用受试者工作特征(ROC)曲线验证列线图预测模型.结果 纳入的306例前置胎盘妊娠孕妇患者中,单纯前置胎盘组211例,占比68.95％,前置胎盘伴胎盘植入组95例,占比31.05％.单因素分析结果显示,两组患者年龄、人工流产次数、孕次、剖宫产次数、有无剖宫产分娩史、有无宫腔操作史、前置胎盘类型、胎盘厚度方面比较,差异均有统计学意义(P＜0.05).年龄≥35岁、人工流产次数≥2次、有剖宫产分娩史、前置胎盘类型(完整性)、胎盘厚度≥50 mm是前置胎盘伴胎盘植入的独立危险因素(P＜0.05).基于上述因素构建前置胎盘伴胎盘植入列线图,利用ROC曲线对列线图预测模型进行验证,曲线下面积值为0.864(95％CI:0.821～0.908),H-L检验结果显示,该模型具有较好的拟合度(x2=1.136,P=0.793).结论 年龄、人工流产次数、前置胎盘类型、有剖宫产分娩史、胎盘厚度是前置胎盘伴胎盘植入的独立危险因素.据此构建的列线图预测模型具有良好的特异度和敏感度,可用于临床指导,制定最佳的临床干预计划,对于改善母婴结局具有重要意义.
Abstract
Objective To investigate the risk factors of placenta previa with placenta implantation and construct a nomogram prediction model.Methods Three hundred and six pregnant women with placenta previa who were admitted to The Affiliated Obstetrics and Gynecology Hospital of Tongji University from May 2020 to January 2023 were retrospectively selected as the study objects.According to the presence or ab-sence of placenta implantation,they were divided into the simple placenta previa group and the placenta previa with placenta implantation group.The general information of the research subjects(age,body mass index,number of pregnancies,number of induced abortions,number of cesarean sections,original surgical hospital level),previous pregnancy history(cesarean section delivery history,uterine cavity operation history,history of fetal preservation),clinical manifestations during pregnancy(whether vaginal bleeding occurred before delivery),and ultrasound examination re-sults(type of placenta previa,placental thickness)were collected.Logistic regression was used to analyze the risk factors of placenta previa with placenta implantation in patients,and the prediction model based on the risk factors was constructed and verified using receiver operating character-istic(ROC)curve.Results Among the 306 pregnant women with placenta previa,there were 211 cases in the simple placenta previa group,ac-counting for 68.95％,and 95 cases of placenta previa with placental implantation group,accounting for 31.05％.The results of univariate analy-sis showed that there were statistically significant differences between the two groups in terms of age,number of induced abortions,number of preg-nancies,number of cesarean sections,history of cesarean delivery,history of uterine cavity operation,type of placenta previa,and placental thick-ness(P＜0.05).Age≥35 years old,number of induced abortion ≥ 2,history of cesarean delivery,placenta previa type(integrity),and pla-centa thickness≥50 mm were independent risk factors for placenta previa with placenta implantation(P＜0.05).Based on the above factors,a nomogram of placenta previa with placenta implantation was constructed.ROC curve was used to verify the nomogram prediction model.Area under the curve under the curve was 0.864(95％CI:0.821-0.908).The results of H-L test showed that the model had a good fit(x2=1.136,P=0.793).Conclusion Age,number of induced abortions,type of placenta previa,history of cesarean section delivery,and placental thickness are independent risk factors for placenta previa with placental implantation.The prediction model based on this nomogram has good specificity and sensitivity,and can be used for clinical guidance and formulation of the best clinical intervention plan,which is of great significance for improving maternal and infant outcomes.
基金项目
上海市"科技创新行动计划"自然科学基金项目辅助项目(20ZR1443900 2020)
NETL
NSTL
维普
万方数据