Analysis Of Risk Factors And Construction Of Risk Assessment Model For Cerebral Injury Caused By Neonatal Asphyxia

Objective:To identify risk factors that lead to brain injury in asphyxia infants by comparing the clinical manifestations and laboratory indicators between infants with brain injury and those without brain injury.Then conducting single-factor and multiple-factor logistic regression analyses to explore the predictive value of the risk factors in the occurrence of brain injury,and establishing a column chart prediction model to predict and facilitate early clinical intervention to prevent serious neurological sequelae.Methods:A total of 163 cases of perinatal asphyxia infants admitted to our NICU for hospitalization from January 2018 to October 2022 were retrospectively included and divided into cerebral injury group and non-cerebral injury group.The general data,clinical manifestations,relevant clinical indicators,and laboratory indicators of the two groups of infants were compared,and the possible risk factors for cerebral injury in asphyxia infants were obtained by single-factor analysis.The indicators with significant differences were further analyzed by multiple-factor logistic regression to obtain independent risk factors for brain injury in asphyxia infants.ROC curve was drawn to evaluate the predictive efficacy,and a column chart prediction model was established with a calibration curve.Results:（1）A total of 163 cases of neonatal asphyxia infants were included in this study,among which 90 cases had cerebral injury and 73 cases did not,with a cerebral injury incidence rate of 55.2%.（2）In this study,full-term infants were more likely to have cerebral injury than premature infants（P<0.05）.Severe asphyxia infants were more likely to have cerebral injury than mild asphyxia infants（P<0.05）.The clinical manifestations of the cerebral injury group were also more severe than those of the non-cerebral injury group（P<0.05）.（3）The number of infants in the cerebral injury group whose mothers had two or less births,amniotic fluid contamination,intrauterine distress,abnormal umbilical cord,intrauterine infection,maternal age≥35years,maternal anemia during pregnancy,and postnatal fluid restriction and blood transfusion were more than those in the non-cerebral injury group.（P<0.05）.（4）The neuronal enolase（NSE）,PCO₂,LAC,WBC,LYM,NEUT,nucleated red cell count（NRBC）,nucleated red cell/100 leukocyte（NRBC%）,PCT,IL6,AST,AG,GLU,LDH,HBDH and other indexes in cerebral injury group were significantly higher than those in the non-cerebral injury group（P<0.05）.However,the values of SBE,HC0₃^-and PH in the cerebral injury group were lower than those in the non-cerebral injury group（P<0.05）.（5）Binary logistic regression analysis indicated that full-term birth,Apgar score≤3 at 1 minute,elevated NRBC,elevated NSE,and decreased PH were all independent risk factors for cerebral injury in asphyxia infants（P<0.05）.（6）ROC curve analysis suggested that the optimal cut-off values for predicting cerebral injury in asphyxia infants were 1.371×10⁹/L for NRBC,7.129 for PH,and 68.905ng/ml for NSE.（7）Based on the results of the multiple-factor logistic regression analysis,a column chart prediction model was developed,and the calibration plot showed that the predicted curve and the 45°diagonal line representing the actual values were almost overlapped,indicating a good fit and high accuracy of the prediction model.Conclusion:（1）Gestational age,Apgar score≤3 at 1 minute,elevated NRBC,elevated NSE,and decreased PH are independent risk factors for cerebral injury in asphyxia infants.The combined use of these factors has a higher clinical value in predicting brain injury in asphyxia infants.（2）The prediction model of the column graph drawn by independent risk factors was obtained by multiple-factor Logistic regression analysis,which has high prediction accuracy,and can provide references for pediatricians in early detection of cerebral injury in asphyxia infants.