| Background and purposeBronchopulmonary dysplasia(BPD)is one of the most serious complications in premature infants.With the improvement of neonatal intensive care,the survival rate of extremely low/very low birth weight infants(ELBWs/VLBWs)increased significantly,and the incidence of BPD gradually increased.At present,there is no specific drug treatment for BPD.Once it occurs,it will not only lead to poor growth and high incidence of long-term respiratory diseases,but also increase the risk of cardiovascular disease and nervous system developmental disorders,and decrease the long-term quality of life.Compared with mild BPD,children with moderate-to-severe BPD have longer hospital stay,higher mortality,and may be accompanied by lifelong dysfunction.Early positive pressure ventilation or selective use of hormones in infants with moderate-to-severe BPD can reduce the occurrence of serious complications to a certain extent,so it is of great significance to explore the methods of early screening and prevention of moderate-to-severe BPD.There is some evidence that the level of serum N-terminal pro-brain natriuretic peptide(NT-proBNP)is an independent risk factor for predicting BPD,so we speculated that this index may be an early marker of severe BPD.The purpose of this study was to explore the clinical value of serum NT-proBNP level in predicting moderate-to-severe BPD/death in premature infants at different times of hospitalization,and to establish an effective clinical predictive model.MethodsThis retrospective study included premature infants with gestational age ≥32 weeks and birth weight ≤1500g delivered in the obstetrics department of the first affiliated Hospital of Zhengzhou University and admitted to the neonatal intensive care unit from November 1,2016 to February 28,2021.A series of serum NT-proBNP levels were collected on the 1st,3rd,7th,14th,21st and 28th day after birth.The correlation between NT-proBNP level and moderate-to-severe BPD/death was evaluated.Receiver operating characteristic(ROC)curve analysis was used to evaluate the prediction ability.Then,we used multivariable logistic regression to build the prediction model.Finally,the nomogram of the prediction model was established,and the calibration curve was used to evaluate the performance.Results1.A total of 556 preterm infants were included,among whom 229 developed BPD(mild:n=109;moderate:n=68;severe:n=52)and 18 died.The incidence of BPD was 41.2%,and the incidence of moderate-to-severe BPD/death was 24.8%.2.The NT-proBNP level in the moderate-to-severe BPD/death group was significantly higher than that in the no-to-mild BPD group from the 3rd to 28th day(P<0.001).When the natural logarithm of the serum NT-proBNP level increased by 1 unit at day 7(±2 days)of life,the risk of moderate-to-severe BPD/death was the highest(OR=3.753;95%CI:2.984～4.720,P<0.001),and ROC analysis identified an optimal cutoff point of 3360ng/L(sensitivity:80.0%;specificity:86.2%;AUC:0.861).3.After the confounding factors were corrected by multivariate Logistic regression,the level of NT-proBNP at day 7(±2 days)of life still had important predictive value for the development of moderate-to-severe BPD/death,significantly improving the predictive ability of the model.4.The factors included in the final model included gestational age,sex,intrauterine growth restriction,days of mechanical ventilation within one week after birth,type of respiratory support and NT-proBNP level on the 7th day after birth.The area under the ROC curve of the model was 0.906,which had high sensitivity and specificity.Conclusions1.The level of NT-proBNP at day 7(±2 days)of life can be used as an early promising biomarker for VLBWs to develop moderate-to-severe BPD/death.2.We constructed a nomogram that can predict the risk of moderate-to-severe BPD/death early in life,which is helpful for clinicians to identify high-risk populations early and intervene. |
PDF Full Text Request