Construction of Prediction Model of Neonatal Respiratory Distress Syn-drome Based on the Clinical Characteristics of Newborns and Pregnant Women

QU Haixin; ZHANG Yajing; GUO Weiping; YUAN Erwei; WANG Lingling; LI Kun

doi:10.3969/j.issn.1002-3429.2024.08.013

Abstract：Objective To construct a prediction model of neonatal respiratory distress syndrome(NRDS)based on the clinical characteristics of newborns and pregnant women.Methods In total,102 children with NRDS admitted from Jan-uary 2020 to January 2023 were included in the NRDS group,and another 150 healthy newborns were included in the non-NRDS group during the same period.Clinical data were retrospectively collected,and the conditions of newborns and pregnant women in the two groups were analyzed.The risk factors of NRDS were analyzed by multivariate Logistic regression,the pre-diction model was established,and receiver operating characteristic(ROC)curve was drawn to analyze the predictive value of the prediction model for NRDS.Results Compared with the non-NRDS group,the proportion of the gestational age＜34 weeks,birth weight＜2 kg,delivery mode of cesarean section,fetal distress,premature rupture of membranes,acidosis was higher in the NRDS group(P＜0.05,P＜0.01).Compared with the non-NRDS group,the pregnant women in the NRDS group had higher proportion of≥35 years of age,first delivery,gestational hypertension,gestational diabetes,com-bined reproductive tract infection,prenatal examination less than 9 times,but lower levels of serum 25-hydroxyvitamin D(25-OH-D)(P＜0.05,P＜0.01).Gestational age＜34 weeks,birth weight＜2 kg,delivery mode of cesarean section,fetal distress,maternal age≥35 years,gestational diabetes,prenatal examination less than 9 times were independent risk factors for NRDS(P＜0.05,P＜0.01),while the increase of serum level of 25-OH-D in pregnant women was a protective factor(P＜0.01).According to the above results,a prediction model was constructed and the ROC curve was drawn.The results showed that when Logit(P)was＞18.570,the area under the curve for predicting the occurrence of NRDS was0.893,the diagnostic sensitivity was 81.37%,and the specificity was82.00%.Conclusion Gestational age＜34 weeks,birth weight＜2 kg,delivery mode of cesarean section,fetal distress,maternal age≥35 years,gestational diabetes,and prenatal exami-nation less than 9 times were independent risk factors for NRDS,whereas the elevated serum level of 25-OH-D of pregnant women was protective factor.Therefore,the predictive model had high value for predicting NRDS,which could provide a basis for screening high-risk children and formulating clinical intervention strategies.