Establishment And Efficiency Analysis Of The Risk Predictive Model To Predict The Development Of Bronchopulmonary Dysplasia

Objective: We establish an early predictive model of infants with birth weight less than 1500 g to predict the risk to develop for BPD,in order to provide effective guidance to prevent the occurrence of BPD and improve the prognosis of BPD.Methods: We collected data from children's hospital of Chongqing Medical University from January 1th 2014 to May31 th 2016,331 newborns with the birth weight less than 1500 g were retrospectively analyzed,there are 86 cases who needed oxygen until 28 days after birth,named BPD group,while 225 cases over 28 days after birth without oxygen,named no-BPD group,.(1)All the cases were grouped into 3 groups: below1000 g,1000g-1200 g,1201g-1499 g group,analysis the incidence of BPD in every group;(2)we randomly chose 86 cases from the no-BPD group as control group.Compared all the clinical characteristics in the two groups,the clinical characteristics,including the following aspects: antenatal features: mother's age,preeclampsia,GDM,chorioamnionitis,oligohydramnios,PPROM,multiple pregnancies,Steroid administration,IUGR,fetal distress;natal features: Delivery route(cesarean section),Perinatal asphyxia,Delivery room resuscitation,gender(male),gestational age,birth weight;postnatal features: NRDS,PDA,PH,intubation,IVH,Sputum culture,sepsis,pulmonary hemorrhage,Maximum amount of fluid administered during the first3d;Laboratory findings: anemia,Hypoglycemia,Hypocalcemia.(3)The clinical characteristics of was statistically significant analyzed again,aimed to find out the independent risk factors influencing the BPD occur,and to establish a risk assessment model,then evaluation.(4)From June 1th,2016 to December 31 th,2016,61 cases were Validated in NICU with prospective score.(5)All the results of the clinical characteristics were disposed by SPSS17.0 analysis of measurement data,and tried to establish a risk assessment model on the clinical characteristics of statistical significance analysis,then evaluation.Results: The mortality rates of newborns of the Birth weight <1000g,1000g-1200 g,1200g-1499 g,were 17.4%,8.8% and 4.8%.With the increase of birth weight,the incidence of BPD and neonatal mortality are significantly reduced.Chi-square tests showed: Delivery route(cesarean section),Perinatal asphyxia,Delivery room resuscitation,gender(male),NRDS,intubation,PDA,PH,IVH,Sputum culture,sepsis,pulmonary hemorrhage,Hypocalcemia in case group were higher than those in control group.But there were no differences among mother's age,preeclampsia,gestational diabetes,chorioamnionitis,PROM,multiple pregnancies,Steroid administration,IUGR,fetal distress,Maximum amount of fluid administered during the first3 d,Hypocalcemia between the two groups.Multiariate Logistic regression analysis results: neonatal asphyxia,gender(male),Sputum culture,sepsis,PH were the independent risk factors lead to BPD.According to the Logistic model score,< 6.712 is divided into low risk,and > 6.712 divided into high risk.Newborns from June 1th 2016 to December 31th2016 61 cases were brought into test group.Both in the test group,The higher the scores were,the higher the incidence rates of BPD in model group and test group were;The incidence rate of BPD was significant different between high score group and low score group(P < 0.05),and the areas under ROC curve in test group were 0.87,respectively.Conclusion: the risk prediction model can accurately detect infants with birth weight below 1500 g in children with BPD may occur,thus guiding medication,take intervention measures of BPD occur as soon as possible.