硬膜外分娩镇痛产妇产间发热风险预测模型的构建与验证

Construction and validation of a predictive model for postpartum fever risk in women undergoing epidural⁃related maternal fever

目的:构建硬膜外分娩镇痛产妇产间发热(ERMF)风险预测模型并验证效果。方法:采用前瞻性研究方法,选取2022年1月—2022年7月于江苏省某三级甲等妇产医院产房接受硬膜外分娩镇痛的780名产妇为研究对象,根据是否诊断为ERMF分为ERMF组(n=148)与非ERMF组(n=632)。使用Logistic回归分析确定ERMF的独立危险因素,构建风险预测模型。采用受试者工作特征(receiver operating characteristic,ROC)曲线与Hosmer-Lemeshow检验评价该模型的预测效能与拟合优度。选取2022年8月—10月住院的170名产妇作为模型的验证组,对预测模型进行验证。结果:建模组ERMF发生率为19.0%,验证组ERMF发生率20.6%。阴道指检次数、缩宫素使用、产程、胎膜破裂至分娩结束时间、罗哌卡因用量、舒芬太尼用量是ERMF的独立危险因素。预测模型的ROC曲线下面积为0.987,灵敏度为0.912,特异度为0.796,约登指数为0.888。验证组ROC曲线下面积为0.893,灵敏度为0.886,特异度为0.785。结论:该预测模型具有较好的区分度和校准度,对于识别高危ERMF产妇具有一定的预测价值,可以给临床护理工作提供一定的借鉴与参考,从而制定相应的护理措施以减少ERMF风险。

Objective:To construct a risk prediction model of epidural-related maternal fever(ERMF)and verify the effect.Methods:A prospective study was conducted in a Grade-A obstetrics and gynecology hospital in Jiangsu Province.A total of 780 parturients who received epidural analgesia in the delivery room from January 2022 to July 2022 were selected as subjects.According to diagnosis,they were divided into ERMF group(n=148)and non-ERMF group(n=632).The independent risk factors of ERMF were determined by Logistic regression analysis,and the prediction formula was obtained to construct the risk prediction model.Receiver operating characteristic(ROC)and Hosmer-Lemeshow test were used to evaluate the prediction efficiency and goodness of fit of the model.A total of 170 parturients hospitalized from August 2022 to October 2022 were selected as the validation group of the model to verify the prediction model.Results The incidence of ERMF was 19.0%in the modeling group and 20.6%in the validation group.Number of examinations of the vaginal tract,oxytocin use,labor,the time from rupture of membranes to the end of labor,ropivacaine dosage,sufentanil dosage was an independent risk factor of ERMF.The ROC curve of the prediction model was 0.987,the sensitivity was 0.912,the specificity was 0.796,and the Youden index was 0.888.The ROC curve of the verification group was 0.893,the sensitivity was 0.886,the specificity was 0.785.Conclusion:By verifying that the prediction model has good differentiation and calibration degree,it has certain prediction value for identifying high-risk ERMF parities,and can provide reference for clinical nursing work,so as to formulate corresponding nursing measures to reduce the risk of ERMF.