Prediction Of The Incidence Of Preeclampsia In Pregnant Women With Preeclampsia Risk Factors By A Combination Of Multi-indicator Tests In Early Pregnancy

Objective:In this study,pregnant women with risk factors of preeclampsia were selected as research objects,to investigate the predictive effects of mean arterial pressure,placental growth factor and uterine artery pulsation index on the development of pre-eclampsia in pregnant women with different levels of risk factors in early gestation,and to develop a predictive model for early and late pre-eclampsia in this study population,and compare the predictive effects of the models.Methods:(1)This study used a nested case-control study method.A total of 695 pregnant women who had regular obstetric check-ups and delivered at the West Coast Hospital of the Affiliated Hospital of Qingdao University and Huangdao District Maternal and Child Health Hospital from August 2019 to October 2020 were followed up,and 508 pregnant women with risk factors for preeclampsia after screening were included in the analysis.(2)Pregnant women with only intermediate risk factors for preeclampsia(PE)were classified as intermediate risk pregnant women,and those with one or more high risk factor for PE were classified as high risk pregnant women.Among them,388 pregnant women without preeclampsia were included as the control group,non-PE group,while pregnant women with preeclampsia were included as the case group,PE group,including 36 cases of early-onset PE and 84 cases of late-onset PE.(3)The mean arterial pressure(MAP)was measured by standard methods and the serum placental growth factor(PLGF)concentration was measured by fluorescence immunosorbent assay at 11+0-13+6 weeks of gestation,while the bilateral uterine artery pulsation index(UTA-PI)was measured by colourful Doppler ultrasound.Pregnancy status and pregnancy outcome were recorded.(4)The test results of MAP,PLGF and UTA-PI were transformed into the multiple median,Mo M,of the control group(non-PE group)for data processing,and the differences between groups of MAP,PLGF and UTA-PI were compared using analysis of variance by SPSS25.0.Logistic regression was performed with MAP,PLGF,UTA-PI and pre-eclampsia risk factors as independent variables or only MAP,PLGF and UTA-PI as independent variables,and the occurrence of pre-eclampsia as the dependent variable,to establish prediction models for the occurrence of early-onset PE and late-onset PE in pregnant women with different levels of risk,and to draw ROC curves.The area under curve(AUC),sensitivity,specificity,negative predictive value(NPV)and positive predictive value(PPV)were used to evaluate the predictive effect of each prediction model.With Med Calc software,the area under the ROC curve of the prediction models was compared to evaluate the differences in the predictive effects of the models.Results:(1)Comparison of the general clinical information of pregnant women in each group:the proportion of pregnant women with SLE/APS,history of pre-eclampsia or family history of pre-eclampsia was higher in the early-onset PE group than in the non-PE group(P<0.05).The difference between the late-onset PE group and the non-PE group in terms of SLE/APS,history of pre-eclampsia and family history of pre-eclampsia was not significant(P>0.05),while the interval between pregnancies in late-onset PE was longer than in the early-onset PE and non-PE groups(P<0.05).There were no significant differences between the three groups in terms of previous chronic hypertension,diabetes and age and BMI(P>0.05).(2)Comparison of MAP,PLGF and UTA-PI between groups of pregnant women:there were significant differences in MAP,PLGF and UTA-PI between the early-onset PE and non-PE groups and between the late-onset PE and non-PE groups(P<0.05);while the differences in PLGF between the early-onset PE and late-onset PE groups were statistically significant(P<0.05),The differences in MAP and UTA-PI were not significant(P>0.05).(3)The prediction model for the development of early-onset PE in pregnant women at high risk using only MAP,PLGF,and UTA-PI was P=e~Y/(1+e~Y),Y=0.193(MAP)-0.040(PLGF)+3.334(UTA-PI)-16.184,the AUC of 0.945 was higher than that of the combined prediction of two indexes.MAP,PLGF,UTA-PI combined with high risk factor to create a prediction model for the development of early-onset PE in pregnant women at high risk of developing early-onset PE as P=e~Y/(1+e~Y),Y=0.236(MAP)-0.038(PLGF)+2.130(UA-PI)+0.269(history of chronic hypertension)+1.218(history of diabetes)+1.546(history of SLE/APS)+0.464(history of PE)-19.280,and the AUC of 0.949 was higher than that of two indicators combined with risk factors.The predictive validity of the model for early-onset PE using only the three indices(AUC=0.945)did not differ statistically significance from that of the model when high risk factors were combined(AUC=0.949)(P>0.05).(4)The prediction model for the development of late-onset PE in pregnant women at high risk using only MAP,PLGF,and UTA-PI was P=e~Y/(1+e~Y),Y=0.260(MAP)-0.030(PLGF)+7.155(UTA-PI)-28.974,the AUC of 0.944 was higher than that of the combined prediction of two indexes.MAP,PLGF,UTA-PI combined with high risk factor to create a prediction model for the development of late-onset PE in pregnant women at high risk of developing late-onset PE as P=e~Y/(1+e~Y),Y=0.221(MAP)-0.032(PLGF)+7.174(UTA-PI)+1.645(history of chronic hypertension)+0.997(history of diabetes)+1.306(history of SLE/APS)+1.368(history of PE)-27.034,and the AUC of 0.950 was higher than that of two indicators combined with risk factors.The predictive validity of the model for late-onset PE using only the three indices(AUC=0.944)did not differ statistically significance from that of the model when high risk factors were combined(AUC=0.950)(P>0.05).(5)The prediction model for the development of early-onset PE in pregnant women at intermediate risk using only MAP,PLGF,and UTA-PI was P=e~Y/(1+e~Y),Y=0.134(MAP)-0.023(PLGF)+3.384(UTA-PI)-14.837,the AUC of 0.911 was higher than that of the combined prediction of two indexes.MAP,PLGF,UTA-PI combined with intermediate risk factor to create a predicti on model for the development of early-onset PE in pregnant women at intermediate r isk of developing early-onset PE as P=e~Y/(1+e~Y),Y=0.272(MAP)-0.038(PLGF)+4.543(UTA-PI)+1.629(age≥35 years old)+0.766(BMI≥28Kg/m~2)+2.257(family history of P E)+0.083(primiparity)+2.516(pregnancy interval≥10years)-31.380,and the AUC of 0.939 was higher than that of two indicators combined with risk factors.The predictiv e validity of the model for early-onset PE using only the three indices(AUC=0.911)did not differ statistically significance from that of the model when intermediate risk factors were combined(AUC=0.939)(P>0.05).(6)The prediction model for the development of late-onset PE in pregnant women at intermediate risk using only MAP,PLGF,and UTA-PI was P=e~Y/(1+e~Y),Y=0.075(MAP)-0.013(PLGF)+3.676(UTA-PI)-10.555,the AUC of 0.821 was higher than that of the combined prediction of two indexes.MAP,PLGF combined with intermediate risk factor to create a prediction model for the development of late-onset PE in pregnant women at intermediate risk of deve loping late-onset PE as P=e~Y/(1+e~Y),Y=0.127(MAP)-0.029(PLGF)+1.214(age≥35years old)+3.045(BMI≥28Kg/m~2)+2.266(family history of PE)+0.284(primiparity)+3.593(pregnancy interval≥10years)-13.721,and the AUC of 0.924 was higher than that of three indications combined risk factors(AUC=0.867)and MAP and UTA-PI combi ned risk factors(AUC=0.865).However,when MAP and PLGF were only used in co mbination with moderate risk factors,the predictive efficiency(AUC=0.924)of the pr ediction model was higher than that of the three indicators combined only(AUC=0.821)(P<0.05).Conclusions:(1)MAP,PLGF and UTA-PI in early pregnancy are predictive of the occurrence of pre-eclampsia in pregnant women with different levels of risk factors,and have some significance in guiding the precise prevention of pre-eclampsia.(2)Regarding whether women with high risk factors of preeclampsia develop early onset PE or late onset PE,and whether women with intermediate risk factors of preeclampsia develop early onset PE,namely MAP,PLGF and UTA-PI,is not significantly different from that of the prediction model when risk factors are combined,and the method is simple and more convenient for clinical analysis and evaluation.(3)For pregnant women with intermediate risk factor of preeclampsia,the predictive model established by MAP and PLGF combined with various intermediate risk factors also has a good predictive efficiency for late onset PE.