Analyzing Risk Factors Of Placental Abruption And Establishing A Prediction Model

Objective To analyze the risk factors of placental abruption and establish “a prediction model”,to illustrate the cause of placental abruption and provide the basis to assess the risks of placental abruption.Methods A retrospective analysis was made on the maternal clinical information in Fujian Maternity and children Health Hospital of maternal hospital delivery from January 1 to December 31,2017.Case-controlled study with matched pairs(1:1)was conducted.Patients who confirmed by clinical or pathological diagnosis with placental abruption were selected as study group.The control group consisted of health patients at the same period.Single factor analysis was used for screening risk factors of placental abruption,multifactor logistic regression analysis was used to screen independent risk factors.According to the Odds Ratios(OR)and normal international range of hysiological indicators during pregnancy,“a prediction model of placental abruption”was established.The score value of each patient was plotted by receiver-operating characteristics(ROC),and the predictive efficiency of the “a prediction model of placental abruption” was evaluated according to the area under the curve(AUC)under the ROC curve.All data analysis was performed using SPSS18.0statistical software.Results1.In this study,the incidence rate of placental abruption was 1.79%.Women classified with a class II placental abruption accounted for the highest proportion(41.25%)in all placental abruption,Women classified with a class III placental abruption accounted for the lowest proportion(2.31%)in all placental abruption.2.Compared with the control group,postpartum hemorrhage(PPH),blood transfusion,stayed in Intensive care unit(ICU),duration of hospital stay and hospitalized expense had increased,and the difference was statistically significant(P<0.05).At fetal outcomes,the study group had a high rate in preterm neonates,small for gestation age(SGA),neonatal asphyxia,neonatal pneumonia,respiratoryfailure,neonatal anemia,abnormal coagulant function,hypoproteinemia,neonatal ischemia anoxic myocardial damage,Neonatal Intensive Care Unit(NICU),the difference was statistically significant(P <0.05).3.Single factor analysis of placental abruption showed that there were significant differences in advanced maternal age,multigravida,irregular prenatal care,uterine malformation,assisted reproductive techniques(ART),twin pregnancy,hypertensive disorder complicating pregnancy(HDCP),fetal growth restriction(FGR),anemia,premature rupture of membrane(PROM),excessive short cords,increased placental thickness >5.0centimeter(cm).4.The multifactor logistic regression analysis finally had 7 entry regression model,which were advanced maternal age,multigravida,irregular prenatal care,HDCP,PROM,FGR and anemia.5.The AUC of ROC curve "a prediction model of placental abruption " was 0.777.when the total score reached 4 points,the sensitivity of the placental abruption was78.5%,and the specificity was 61.1%.The best predictive cutting point is 5,with a sensitivity of 67.3%,a specificity of 74.3%.Conclusion1.The maternal and fetal prognosis of placental abruption is poor,and the risk of PPH,blood transfusion and admission to ICU increased,the cost of treatment increased and the length of hospital stay is prolonged.There is an increase in the risk of SGA,premature neonatal,neonatal asphyxia and NICU.2.Placental abruption is closely related to advanced maternal age,multigravida,irregular prenatal care,HDCP,PROM,FGR and anemia.3."A prediction model of placental abruption" has a high sensitivity and specificity to the identification of the placental abruption.The prediction total score ? 5 points is closely related to high risk of placental abruption,and such patients should be required closely monitoring.4."A prediction model of placental abruption" is a simple and practical method to identify the high-risk patients with placental abruption,and has a certain predictive effect on the placental abruption.