Development And Validation Of A Risk Scoring Tool For Bronchopulmonary Dysplasia In Preterm Infants

Objective:This study aimd to screen out the risk factors of bronchopulmonary dysplasia（BPD）in preterm infants based on meta-analysis,established a simple and easy-to-use BPD risk scoring tool for preterm infants,and conducted external validation,providing an effective tool for early diagnosis of high-risk children with BPD,and provide correct and efficient decision-making support for early and timely adoption of appropriate treatment and nursing intervention.Methods:（1）Meta-analysis:According to the established search strategy and inclusion and exclusion criteria,meta-analysis of the literature on risk factors for BPD in preterm infants was performed to identify statistically significant risk factors and corresponding odds ratio（OR）.（2）Development of the risk scoring tool for BPD in preterm infants:According to the screening results of meta-analysis,the OR value and 95%confidence interval（CI）of risk factors for early predicted were selected,and the corresponding regression coefficients were calculated,so as to construct the logistic regression risk prediction model,and then the risk scoring tool was established by scoring the weight of risk factors in the model.（3）Validation of the risk scoring tool for BPD in preterm infants:A cohort of 535preterm infants hospitalized in the Department of Neonatology,the First Affiliated Hospital of Anhui Medical University from June 1,2017 to June 1,2022 were selected for external validation of the scoring tool.Draw the receiver operating characteristic curve（ROC）and evaluate the predictive power of the tool by the area under the curve（AUC）,the calibration curve was plotted using 1000 repeated sampling and the Hosmer-Lemeshow goodness-of-fit test were used to assess the calibration capability of the tool.The clinical usefulness of the tool was assessed using decision curve analysis（DCA）.The optimal cut-offs were calculated according to the sensitivity and specificity of different cut-off points.Preterm infants were classified into risk stratifications and further integrated into the BPD risk scoring tool for preterm infants.Results:（1）This meta-analysis finally included 58 articles,and about 83,034preterm infants with gestational age<32 weeks and/or birth weight<1500 g were screened out.Meta-analysis results showed that there were 9 risk factors with statistically significant（P<0.05）,as follows:Chorioamnionitis（OR=2.82,95%CI[1.92-4.15]）,Sex（OR=1.50,95%CI[1.38-1.63]）,Small for gestational age（SGA）（OR=2.90,95%CI[1.94-4.35]）,Delivery room intubation（OR=2.66,95%CI[2.02-3.50]）,Mechanical ventilation（OR=3.28,95%CI[2.14-5.03]）,Surfactant（OR=3.42,95%CI[2.43-4.83]）,Respiratory distress syndrome（RDS）（OR=4.47,95%CI[2.84-7.04]）,Patent ductus arteriosus（PDA）（OR=3.09,95%CI[2.31-4.13]）and Sepsis（OR=2.07,95%CI[1.67-2.57]）.There were 3protective factors with statistically significant（P<0.05）,as follows:Gestational age（OR=0.64,95%CI[0.58-0.70]）,Birth weight（OR=0.80,95%CI[0.77-0.83]）and 5 min Apgar score（OR=0.71,95%CI[0.64-0.78]）.（2）Some early predictors were selected to be included in the BPD risk prediction model for preterm infants.The 9 predictors of this model include Chorioamnionitis,Gestational age,Birth weight,Sex,SGA,5 min Apgar score,Delivery room intubation,Surfactant and RDS.Based on the weight of each risk factor,we transformed it into a simple clinical scoring system,with the total score ranging from 0 to 64.The scores for inclusion of risk factors were as follows:Chorioamnionitis（no=0,yes=5）,Gestational age(≥32=0,31-31⁺⁶=2,30-30⁺⁶=4,29-29⁺⁶=6,28-28⁺⁶=8,27-27⁺⁶=10,26-26⁺⁶=12,25-25⁺⁶=14,24-24⁺⁶=16,<24=18),Birth weight（≥1500=0,1400-1499=1,1300-1399=2,1200-1299=3,1100-1199=4,1000-1099=5,900-999=6,800-899=7,700-799=8,600-699=9,500-599=10,<500=11）,Sex（female=0,male=2）,SGA（no=0,yes=6）,5 min Apgar score（≥8=0,7=1,6=2,5=3,4=4,3=5,≤2=6）,Delivery room intubation（no=0,yes=4）,Surfactant（no=0,yes=5）,RDS（no=0,yes=7）.（3）The external validation results showed that the tool had good discrimination（AUC=0.907）,the Hosmer-Lemeshow goodness-of-fit test showed a good fit（P=0.3572）,and the calibration curve and decision curve analysis showed that the tool had significant consistency and clinical usefulness.When the optimal cut-off value was 25.5,the sensitivity and the specificity was 0.897 and 0.873,respectively,and the resulting risk scoring tool could classify preterm infants into low-risk,intermediate-risk and high-risk groups.Conclusion:This study developed and validated an effective predictive scoring tool for BPD in preterm infants based on meta-analysis.The tool is easy to promote and practical,and is suitable for preterm infants with a gestational age of<32 weeks and/or birth weight<1500 g.This simple tool may play an important role in establishing early screening strategies for BPD in preterm infants and may guide early,timely treatment and care interventions.This has far-reaching implications for reducing the incidence of BPD and improving the prognosis of children.