Development And Validation Of A Nomogram To Predict The Risk Of Neonatal White Matter Damage

Background and Purpose:In recent years,with the progress of neonatal critical medical treatment technology,neonates’ survival rate with serious diseases has increased significantly;At the same time,there is also an increase in sequelae affecting physical and mental development among these surviving infants.White matter damage(WMD)is the most common and the leading cause of permanent motor and cognitive impairment.Currently,there is no unified view on the treatment of WMD.Consequently,early prediction of the risk of WMD and early prevention of the disease are of great importance.The purpose of this study is to develop a model to predict the risk of neonatal WMD based on perinatal characteristics.Materials and Methods:We collected data of 1,733 infants hospitalized at the Department of Neonatology at The First Affiliated Hospital of Zhengzhou University from 2017 to 2020.Infants were randomly assigned to training(n=1,216)or validation(n=517)cohorts at a ratio of 7:3.Multivariate logistic regression and least absolute shrinkage and selection operator(LASSO)regression analyses were used to establish a risk prediction model on the training cohort data.The predictive accuracy of the model was verified in the validation cohort.The C index of the training cohort was calculated to evaluate the discriminant and predictive ability of the model,and the validation cohort data was utilized for validation.Calibration curves were selected to assess the model’s consistency,and ROC curves were used to calculate the sensitivity and specificity of the model,and risk cutoff values were developed for the high-risk and low-risk populations.Ultimately,the clinical decision curve was analyzed to calculate the clinical benefit of the model.Results:1.Selection of perinatal factors.According to LASSO regression results,the optimal λ was 0.0256,and log(λ)=-3.6652.Five best predictors of WMD were selected,including gestational age,birth weight,fetal distress,premature rupture of membrane,and prenatal corticosteroid use.2.Development and verification of the nomogramThese factors screened by LASSO were analyzed by logistic regression analysis,and there were 4 factors with statistically significant differences(p<0.05).The above four factors were adopted to construct the nomogram.In the training and validation cohorts,the C-index of the nomogram was 0.898(95%confidence interval:0.87450.9215)and 0.887(95%confidence interval:0.8478-0.9262),respectively.The nomogram’s ROC display model had high accuracy,with an area under the curve of 0.8979,a sensitivity of 82.5%,a specificity of 81.7%,and a Youden index of 0.099.3.Clinical practicability of the nomogramDecision curve analysis showed that 1-61%of newborns could benefit from using this prediction model.Conclusion:1.Gestational age,fetal distress,premature rupture of membranes and prenatal use of corticosteroids were independent influencing factors of neonatal WMD.2.The first risk prediction model of neonatal WMD constructed according to the above four indicators has been tested with reasonable accuracy.