摘要
目的基于孕期数据探讨产后盆底功能障碍性疾病(PFD)的危险因素并建立Logistic预测模型。方法采用现况调查和回顾性队列研究方法,收集本院2022年5月—2023年5月常规产检并足月分娩以及产后常规进行盆底功能测定的221例产妇作为研究对象,根据产后盆底超声测定分为PFD组129例和非PFD组92例。回顾性分析两组患者孕期临床资料(包括年龄,孕前体重,产次,吸烟史,饮酒史,分娩方式,各个产程时间,妊娠合并症,胎儿娩出时体重,PFD家族史,长期便秘以及长期咳嗽等),采用单因素分析、Lasso回归分析、Logistic多因素回归分析,绘制列线图(nomogram)建立Logistic预测模型,筛选出导致产后盆底功能障碍性疾病的独立危险因素。结果单因素分析表明两组年龄>35岁、产次≥3次、孕前BMI≥25 kg/m^(2),本次分娩是否剖宫产,第二产程时间≥1 h,PFD家族史,新生儿体重≥4000 g,合并长期咳嗽、长期便秘差异有统计学意义(P<0.05),两组在吸烟史,饮酒史,妊娠合并高血压疾病,妊娠合并糖尿病,第一产程时间,第三产程时间差异无统计学意义(P>0.05)。对单因素分析有差异的9个因素进行Lasso回归分析,将非零特征预测因子纳入Logistic多因素回归分析,发现年龄>35岁、产次≥3次、孕前BMI≥25 kg/m^(2),第二产程时间≥1 h,新生儿体重≥4000 g,PFD家族史是产后PFD独立危险因素(P<0.05),而本次分娩为剖宫产为产后PFD的保护因素(P<0.05)。基于Logistic多因素回归分析结果建立nomogram图Logistic预测模型,并对预测模型进行内部交叉验证,结果显示校准曲线、ROC曲线、DCA曲线、临床影响曲线均提示模型精准度较好。结论本次研究建立的nomogram图Logistic预测模型可较好地预测产后PFD的发生,对临床上预防产后PFD以及筛选产后PFD的高危人群具有较好应用价值。
Objective To explore the risk factors for postpartum pelvic floor dysfunction(PFD)based on pregnancy data and establish a Logistic prediction model.Methods A total of 221 parturients who underwent routine labor examination,full-term delivery and routine pelvic floor function measurement in our hospital from May 2022 to May 2023 were collected as the study objects,with the method of status investigation and retrospective cohort study.According to the ultrasonic measurement of postpartum pelvic floor,129 cases were divided into PFD group and 92 cases were non-PFD group.Clinical data during pregnancy(including age,pre-pregnancy weight,delivery time,smoking history,drinking history,mode of delivery,duration of each labor stage,pregnancy complications,fetal birth weight,family history of PFD,chronic constipation and chronic cough,etc.)of the two groups were retrospectively analyzed.By using univariate analysis,Lasso regression analysis and Logistic multifactor regression analysis,a nomogram was constructed to establish a Logistic prediction model and screen out independent risk factors for postpartum pelvic floor dysfunction.Results Univariate analysis showed that there were statistically significant differences between the two groups in terms of age>35 years,parity≥3 times,pre pregnancy BMI≥25 kg/m^(2),whether the delivery was cesarean section,second stage of labor≥1 hour,PFD family history,newborn weight≥4000 g,combined with long-term cough and long-term constipation(P<0.05).There was no significant difference between the two groups in smoking history,drinking history,pregnancy with hypertension,pregnancy with diabetes,the first stage of labor,and the third stage of labor(P>0.05).Lasso regression analysis was conducted on 9 factors with differences in univariate analysis.Non zero feature predictive factors were included in Logistic multiple regression analysis.It was found that age>35 years old,parity≥3 times,pre pregnancy BMI≥25 kg/m^(2),second stage of labor≥1 hour,newborn weight≥4000 g,and family history of PFD were independent risk factors for postpartum PFD(P<0.05).Cesarean section was a protective factor for PFD(P<0.05).Based on the results of Logistic multiple factor regression analysis,a nomogram Logistic prediction model was established,and internal cross validation was performed on the prediction model.The results showed that the calibration curve,ROC curve,DCA curve,and clinical impact curve all indicated good accuracy of the model.Conclusion The nomogram Logistic prediction model established in this study can effectively predict the occurrence of postpartum PFD,and has good application value for clinical prevention of postpartum PFD and screening of high-risk populations for postpartum PFD.
作者
马一虎
张俊茹
马妍
白璐
姚念玲
乔谷媛
马向东
MA Yihu;ZHANG Junru;MA Yan;BAI Lu;YAO Nianling;QIAO Guyuan;MA Xiangdong(Department of Obstetrics and Gynecology,The First Affiliated Hospital of Air Force Military Medical University,Xi'an 710032,China)
出处
《西部医学》
2025年第9期1348-1353,1358,共7页
Medical Journal of West China
基金
陕西省重点研发计划(2022SF-030)。
