Risk factors analysis and predictive modeling of meconium-stained amniotic fluid
ZHAI Ya-li
WANG Hui-ping
GAO Jing-min
MA Ying-li
WANG Yong-juan
LIU Lan
LIU Ye
WANG Yue-lan
Abstract：Objective To explore the risk factors analysis and predictive modeling of meconium-stained amniot-ic fluid.Methods A total of 2 174 pregnant women admitted to our hospital between January 2019 and January 2023 were included in the study.Based on the status of amniotic fluid,they were divided into two groups:the contaminated group(n=1 761)and the non-contaminated group(n=413).Chi-square test,Mann-Whitney U test or t test were used to screen the baseline maternal characteristics by single factor analysis,and multi-factor logistic regression analysis was further used to identify the key factors affecting the amniotic fluid fetal contamination,and then construct the corre-sponding prediction model.Results Logistic regression analysis showed that a total of 9 indicators were significantly cor-related with the occurrence of MSAF(P＜0.05),which were:parity,delivery time,fetal position,gestational diabetes mellitus,abnormal fetal heart monitoring,abnormal labor process,oligohydramnios,delivery mode and gestational week at delivery.A Logistic regression model was constructed,and the model expression was as follows:logit(P)=-0.681-0.240X1-0.306X2-0.109X3+0.653X4+0.482X5+0.336X6+0.412X7+0.605 X8+0.789X9,the meaning of each variable is as follows:X1:parity,X2:birth,X3:Fetal position,X4:gestational diabetes mellitus,X5:abnormal fetal heart monitoring,X6:abnormal labor process,X7:oligohydramnios,X8:mode of delivery,X9:gestational week at deliv-ery.Through ROC curve analysis,the optimal diagnostic critical value of the model was determined to be 0.724,the sen-sitivity of the model was 83.78％,the specificity was 76.38％,and the AUC value was 0.882(95％confidence interval:0.846～0.917),indicating that the model had high predictive value.Conclusion Meconium-stained amniotic fluid is influenced by multiple factors,including parity,birth order,fetal position,gestational diabetes,abnormal fetal heart rate monitoring,labor abnormalities,oligohydramnios,mode of delivery,and gestational age at delivery.The predictive model developed based on these factors demonstrates high predictive accuracy and can effectively forecast the occurrence of meco-nium-stained amniotic fluid in clinical practice,thereby optimizing delivery outcomes.
Keywords：meconium-stained amniotic fluidrisk factorsprediction model
Publication Date：2025-09-15
Online Publishing Date：2025-11-06(First online date of this platform, not the publication date of the document)
Pages：6( 1351-1356 )