硬膜外分娩镇痛产妇绒毛膜羊膜炎相关产时发热预测模型的建立
Construction of a prediction model for intrapartum fever related to chorioamnionitis in parturients undergoing epidural labor analgesia
凌亮 1刘波 2魏大源 3陈本祯 3肖洪权 3张健3
作者信息
- 1. 川北医学院临床医学院,南充 637100
- 2. 成都市锦江区妇幼保健院麻醉科,成都 610000
- 3. 四川省妇幼保健院麻醉科,成都 610000
- 折叠
摘要
目的 建立硬膜外分娩镇痛产妇绒毛膜羊膜炎相关产时发热的预测模型.方法 本研究为回顾性研究.选择四川省妇幼保健院2020年1月至2022年12月接受硬膜外分娩镇痛期间发生发热(腋温≥38 ℃)的产妇纳入建模组,选择四川省妇幼保健院2023年1月至10月接受硬膜外分娩镇痛期间发生发热的产妇纳入为验证组.建模组根据胎盘组织病理学检查结果分为组织学绒毛膜羊膜炎分期≥ Ⅱ期组(HCA≥Ⅱ期组)和组织学绒毛膜羊膜炎分期＜Ⅱ期(HCA＜Ⅱ期组).采用logistic回归分析筛选产妇绒毛膜羊膜炎相关产时发热的独立危险因素,并基于此建立列线图预测模型.通过受试者工作特征曲线下面积验证模型的区分度,通过校准曲线验证模型的一致性,通过决策曲线确定模型的临床有效性;采用验证组数据进一步评估模型.结果 最终建模组纳入产妇308例,验证组99例.多因素logistic回归分析显示,孕周、羊水粪染、C反应蛋白浓度和最高体温是硬膜外分娩镇痛产妇绒毛膜羊膜炎相关产时发热的独立危险因素(P＜0.05).基于此建立列线图风险预测模型,建模组和验证组曲线下面积(95％置信区间)分别为0.844(0.744～0.944)、0.812(0.674～0.950).校准曲线显示模型预测概率与实际诊断概率具有较好的一致性.决策曲线显示预测模型在建模组和验证组的阈值概率分别为10％～98％和10％～78％.结论 基于孕周、CRP浓度、羊水粪染和最高体温成功建立了硬膜外分娩镇痛产妇绒毛膜羊膜炎相关产时发热的列线图预测模型,该模型有良好的预测性能及临床价值.
Abstract
Objective To develop a predictive model for intrapartum fever related to chorioamnion-itis in parturients undergoing epidural labor analgesia.Methods This was a retrospective study.The partu-rients with intrapartum fever(axillary temperature≥38 ℃)who received epidural labor analgesia from Jan-uary 2020 to December 2022 in Sichuan Maternal and Child Health Hospital were selected as model group,and parturients with intrapartum fever who received epidural labor analgesia from January to October 2023 in Sichuan Maternal and Child Health Hospital were selected as validation group.The parturients in model group were divided into histological chorioamnionitis stage ≥ Ⅱ group(HCA ≥ Ⅱ group)and histological chorioamnionitis stage＜Ⅱ group(HCA＜Ⅱ group)according to the results of placental histopathological examination.Logistic regression analysis was used to screen the independent risk factors for intrapartum fever related to chorioamnionitis in parturients,and then a nomogram model was established.The discrimination of the model was verified by the area under the the receiver operating characteristic curve.The consistency of the model was verified by the calibration curve,and the clinical effectiveness of the model was determined by the decision curve.The validation dataset was used to further evaluate the model.Results A total of 308 parturients were finally included in model group and 99 parturients in validation group.Multivariate logistic regression analysis showed that the gestational age,meconium-stained amniotic fluid,c-reactive protein con-centration and maximum body temperature were independent risk factors for intrapartum fever related to cho-rioamnionitis in parturients undergoing epidural labor analgesia(P＜0.05).Based on this,a nomogram risk prediction model was developed.The area under the curve(95％confidence interval)was 0.844(0.744-0.944)in model group and 0.812(0.674-0.950)in validation group.The calibration curve showed that the prediction probability of the model had good consistency with the actual probability of diagnosis.The decision curve showed that the threshold probability of the prediction model in model group and validation group was 10％-98％and 10％-78％,respectively.Conclusions A nomogram prediction model for intrapartum fever related to chorioamnionitis is successfully established based on the gestational age,c-reactive protein con-centration,meconium-stained amniotic fluid and maximum body temperature in parturients undergoing epi-dural labor analgesia.The model has good predictive performance and clinical value.
关键词
分娩疼痛/镇痛,硬膜外/发热/绒毛膜羊膜炎/预测Key words
Labor pain/Analgesia,epidural/Fever/Chorioamnionitis/Prediction引用本文复制引用
出版年
2024
NETL
NSTL
万方数据