Establishment Of The Risk Prediction Scale Of Bilirubin Encephalopathy In Term And Late Preterm Infants

Objective:To analyze the risk factors,clinical manifestations and auxiliary examination of bilirubin encephalopathy(BE)among extreme hyperbilirubinemia(NHB)in term and late preterm infants,and establish a comprehensive,simple,objecctive and effective scoring model for early prediction of bilirubin encephalopathy.Methods:Neonates 35 or more weeks of gestation diagnosed as extreme NHB who were admitted to the neonates department of Bethune First Hospital of Jilin University from January 1st,2012 to December 31st,2019 were selected as research objects.A neurologic evaluation was performed using the bilirubin-induced neurologic dysfunction(BIND)criteria.We defined BE as a total BIND score of 4 to 9 and included all those who were diagnosed as BE in the case group(n=102),while those with a total BIND score of 0 to 3 were randomly included according to the ratio of 1:3in the control group(n=306).A retrospective study of risk factors,clinical characteristics and auxiliary examination related to BE was performed.SPSS 22.0 and Graphpad Prism 8 software were used for single factor correlation analysis and multifactor logistic regression analysis to determine the independent risk factors for BE.We then established a predictive-scoring model using the independent risk factors and used the receiver operator characteristic(ROC)curve to figure out the best diagnostic cutoff point,and calculated the sensitivity and specificity.Result:(1)The results of univariate analysis showed that premature,premature rupture of membranes,spontaneous delivery,advanced maternal age(>age of 35),significantly increased peak total serum bilirubin concentration(TSB)(?35mg/dL),abnormal white blood cell(WBC)count(<5×10~9/L or>21×10~9/L),abnormal hemoglobin concentration(<140g/L or>200g/L),specific signal changes of brain magnetic resonance imaging and hypoalbuminemia were the relevant risk factors of BE(P<0.05).(2)The results of multifactor logistic regression analysis showed that advanced maternal age(OR=2.675,95%CI:1.155-6.197),significantly increased peak TSB(OR=27.324,95%CI:10.754-69.424),abnormal WBC count(OR=8.284,95%CI:3.011-22.789),specific signal changes of brain magnetic resonance imaging(OR=25.597,95%CI:11.700-56.001),hypoalbuminemia(OR=2.176,95%CI:1.072-4.419)were the independent risk factors of BE.(3)A prediction-scoring model was established on the basis of the five independent risk factors mentioned above with a score range of 0 to 10.The ROC curve showed the area under the curve was 0.922(95%CI:0.892-0.947),which was proved to be statistically significant(P<0.05).The greatest Youden index was 0.745when the cutoff point was 3 while the sensitivity and specificity were 85.3%and89.2%respectively.Conclusion:(1)Advanced maternal age(>age of 35),peak TSB?35mg/dL,abnormal WBC count(<5×10~9/L or>21×10~9/L),specific signal changes of brain magnetic resonance imaging and hypoalbuminemia are independent risk factors of BE among extreme NHB in term and late preterm infants.(2)The BE prediction-scoring model is of high value in the prediction of BE caused by extreme NHB in term and late preterm infants.