Construction Of Predictive Model For Expectant Treatment Duration Of Early-onset Severe Preeclampsia And Investigation Of Maternal-infant Outcomes

Objective: To analyze the correlation between expectant management and maternal-infant outcomes of early-onset severe preeclampsia.To explore the independent influencing factors of the duration of expectant management of early-onset severe preeclampsia and to construct a prediction model for the duration of expectant management.Methods: The clinical data of 299 pregnant women with early-onset severe preeclampsia admitted to the Department of Obstetrics,Anhui Provincial Hospital from December 2016 to November 2021 were retrospectively analyzed.After admission,according to whether the fetal lung maturation was completed,they were divided into48 hours delivery group(66 cases)and expectant management group(233 cases).The expectant management group was further divided into short-term expectant management group(expectant management for 3-7 days,n = 130)and long-term expectant management group(expectant management for > 7 days,n = 103)according to the extension of expectant management days.Relevant clinical data were collected for statistical analysis to explore the effect of expectant management on maternal and infant outcomes.R Studio software was used to draw the Nomogram prediction model.The Bootstrap method was used for internal validation,and the ROC curve and calibration curve were used to evaluate the predictive value of the model for long-term expectant management.Results:(1)Compared with the 48 hours delivery group,the expectant treatment group had lower proportion of menpara,postpartum ICU occupancy rate,HELLP syndrome,incidence of blood transfusion and liver damage(P < 0.05).Neonatal NICUs had fewer hospital days,lower incidence of respiratory failure and higher 1-minute Apgar score(P< 0.05).(2)Compared with the short-term expectation group,the incidence of liver damage and postpartum ICU transfer was lower in the long-term expectation group(P <0.05).The incidence of blood glucose at birth,non-invasive ventilation,invasive ventilation,respiratory distress syndrome,pulmonary bronchial dysplasia and intracranial hemorrhage were lower(P < 0.05).(3)Logistic regression analysis showed that gestational age,diastolic blood pressure,blood uric acid concentration and platelet count were independent factors affecting the expected treatment time.(4)Model bias calibration curve and apparent curve fit well.The AUC of ROC curve was 0.724,the sensitivity was 79.4%,and the specificity was 58.9%,which was better than the prediction value of a single index.Conclusion:Gestational age,diastolic blood pressure,uric acid,and platelet count are independent factors affecting the expected treatment duration of women with EOSPE.A Nomogram map model based on these independent factors can be of good value for predicting long-term expected treatment and help obstetricians to evaluate the possibility of long-term expected treatment for women with early-onset severe preeclampsia.Provides a convenient and simple tool to guide clinical decision making;Pregnant women who expect treatment longer than 1 week should be monitored for heart damage,and newborns should be closely monitored for blood sugar to prevent hypoglycemia.