经产妇经阴道分娩产后出血风险列线图模型的建立及验证  

Establishment and validation of a nomogram model for the risk of postpartum hemorrhage in multipara with vaginal delivery

作　　者：李艳鸾 张苑艳 黄启芸 唐丽春[1] LI Yanluan;ZHANG Yuanyan;HUANG Qiyun;TANG Lichun(Delivery Room,Guangxi Zhuang Autonomous Region Maternal and Child Health Hospital,Nanning 530002,Guangxi,China)

机构地区：[1]广西壮族自治区妇幼保健院产房,南宁530002

出　　处：《中国性科学》2024年第2期109-114,共6页Chinese Journal of Human Sexuality

摘　　要：目的分析经产妇经阴道分娩发生产后出血(PPH)的危险因素,并建立预测其发生风险的个体化列线图模型。方法回顾性分析2018年1月至2022年6月在广西壮族自治区妇幼保健院经阴道分娩的7207例经产妇的临床资料,将研究对象随机分为建模组(70%)与验证组(30%),根据是否发生PPH,分为PPH组(n=238)和非PPH组(n=6969)。比较非PPH组和PPH组的临床指标,将建模组数据中差异具有统计学意义的指标作为自变量,是否发生PPH为结局变量,纳入多因素Logistic回归分析,并绘制列线图。用验证组数据进行验证,绘制受试者工作特征(ROC)曲线并计算曲线下面积(AUC)判断模型辨别能力;绘制校准图以判断模型校准能力。结果非PPH组和PPH组年龄(χ^(2)=12.583)、孕期出血史(χ^(2)=10.576)、妊娠期高血压疾病(χ^(2)=18.066)、妊娠期糖尿病(χ^(2)=42.775)、子痫前期(χ^(2)=10.165)、多胎(χ^(2)=16.467)、产程延长(χ^(2)=7.570)、巨大胎儿(χ^(2)=14.306)、凝血功能异常(χ^(2)=62.961)、宫缩乏力(χ^(2)=32.305)、前置胎盘(χ^(2)=20.374)、胎盘早剥(χ^(2)=61.508)、胎盘粘连/滞留(χ^(2)=24.086)比较,差异具有统计学意义(P<0.05)。多因素Logistic回归分析结果显示,孕期出血史(OR=3.003,95%CI:1.393~5.818)、子痫前期(OR=2.751,95%CI:1.347~5.106)、多胎(OR=3.990,95%CI:1.769~8.054)、产程延长(OR=1.861,95%CI:1.140~2.903)、巨大胎儿(OR=2.204,95%CI:1.431~3.287)、凝血功能异常(OR=6.221,95%CI:3.605~10.293)、宫缩乏力(OR=2.553,95%CI:1.778~3.597)、前置胎盘(OR=3.858,95%CI:2.139~6.537)、胎盘早剥(OR=4.431,95%CI:2.818~6.762)、胎盘粘连/滞留(OR=3.324,95%CI:1.969~5.353)是发生PPH的独立危险因素。验证组数据验证后,其ROC曲线的AUC为0.785;校准图其模型校准曲线与标准曲线均接近。结论孕期出血史、子痫前期、多胎、产程延长、巨大胎儿、凝血功能异常、宫缩乏力、前置胎盘、胎盘早剥、胎盘粘连/滞留是经产妇经阴道Objective To analyze the risk factors of postpartum hemorrhage(PPH)in multipara with vaginal delivery,and establish an individualized Nomogram model for predicting the risk.Methods The clinical data of 7,207 cases of multipara with vaginal delivery in Guangxi Zhuang Autonomous Region Maternal and Child Health Hospital from January 2018 to June 2022 were retrospectively analyzed,and the subjects were randomly divided into the modeling group(70%)and the validation group(30%),according to whether PPH occurred,the subjects were divided into PPH group(n=238)and non-PPH group(n=6,969).The clinical indicators of the non-PPH and PPH groups were compared,the indicators in the data of modeling group with statistically significant differences were used as the independent variables,and whether the occurrence of PPH as the outcome variable,were included in the multivariate Logistic regression analysis,and a nomogram was drawn.The data of the validation group was used for validation to draw the receiver operating characteristic(ROC)curve and calculate the area under ROC curve(AUC)to determine the model discrimination ability.The calibration chart was drawn to determine the model calibration ability.Results The age(χ^(2)=12.583),bleeding history during pregnancy(χ^(2)=10.576),gestational hypertension(χ^(2)=18.066),gestational diabetes(χ^(2)=42.775),preeclampsia(χ^(2)=10.165),multiple pregnancy(χ^(2)=16.467),prolonged labor(χ^(2)=7.570),macrosomia(χ^(2)=14.306),abnormal blood coagulation(χ^(2)=62.961),uterine atony(χ^(2)=32.305),placenta previa(χ^(2)=20.374),placental abruption(χ^(2)=61.508),and placental adhesion/retention(χ^(2)=24.086)had significant differences(P<0.05)between PPH and non-PPH group.The results of multivariate Logistic regression analysis showed that history of bleeding during pregnancy(OR=3.003,95%CI:1.393-5.818),preeclampsia(OR=2.751,95%CI:1.347-5.106),multiple pregnancy(OR=3.990,95%CI:1.769-8.054),prolonged labor(OR=1.861,95%CI:1.140-2.903),macrosomia(OR=2.204,95%CI:1.431-3.287),abnormal blood c

关 键 词：经产妇 经阴道分娩 产后出血 临床预测模型 列线图 

分 类 号：R714[医药卫生—妇产科学]