Influencing Factors Analysis And Establishment Of Disease Prediction Model Of Necrotizing Enterocolitis In Preterm Infants

Objective:To analyze the influencing factors of necrotizing enterocolitis(NEC)in preterm infants and the clinical data of blood routine parameters within 24 hours after birth,and construct an objective,simple and meaningful scoring model for early prediction of NEC.Methods:From January 1,2011 to December 31,2020,the case group included 298 preterm infants of NEC conforming to the revised Bell stage for Ⅱ and phase III admitted to the neonatal ward of the First Bethune Hospital of Jilin University.300non-NEC preterm infants who matched the case group with gestational age(GA)±1week,hospitalization time ± 1 week and birth weight ± 200 g were selected as the control group according to the ratio of 1:1.The influencing factors of NEC and the clinical data such as blood routine parameters within 24 hours after birth were retrospectively analyzed.SPSS 23.0 and the rms package of R,version 3.6.1statistical software were used to conduct univariate and multivariate statistical analysis on the influencing factors of NEC,and the independent influencing factors of NEC were identified.Therefore,the prediction nomogram was constructed,and the receiver operator characteristic(ROC)curve and the concordance index(C index)were used to measure the predictive performance of the nomogram.Result:(1)Univariate analysis results showed that patent ductus arteriosus(PDA),Ⅱdegrees or above of intracranial hemorrhage,early onset sepsis(< 3 days after birth),late exposure of antibiotics(> 7 days after birth),pulmonary surfactant,peripherally inserted central catheter(PICC),glucocorticoid(GC),breast milk fortifier,probiotics,red cell transfusion,breast milk,hematocrit(HCT),mean corpuscular volume(MCV),mean platelet volume(MPV),platelet distribution width(PDW)had statistical significance.(P < 0.05)(2)The results of multifactor logistic regression analysis showed thatⅡdegrees or above of intracranial hemorrhage(OR(odds ratio)=1.736,95%CI(confidence interval):1.062-2.839),PICC(OR=1.747,95%CI:1.050-2.906,breast milk fortifier(OR=3.335,95%CI : 1.637-6.796),red cell transfusion(OR=2.827,95%CI :1.494-5.351),HCT(>49.65%)(OR=6.452,95%CI:4.130-10.080),MCV(>114.35 f L)(OR=1.909,95%CI:1.236-2.948),MPV(>10.95 f L)(OR=2.741,95%CI:1.608-4.673)were the independent risk factors of NEC.However,surfactant(OR=0.528,95%CI:0.318-0.877),probiotics(OR=0.348,95%CI:0.229-0.530)and PDW(>11.8f L)(OR=0.514,95%CI:0.304-0.867)were the protective factors of NEC.(3)All associated influencing factors were used to form a NEC risk estimation nomogram.The bootstrap validation method was used to internally validate the resulting model.The nomogram demonstrated good accuracy in predicting the risk of NEC,with a bootstrap-corrected C index of 0.844.The optimal cutoff value of the total nomogram scores was determined to be 171.02.The sensitivity,specificity when used in differentiating the presence from the absence of NEC were 74.7%,80.5%,respectively.Conclusion:(1)Ⅱdegrees or above of intracranial hemorrhage,PICC,breast milk fortifier,red cell transfusion,HCT(>49.65%),MCV(>114.35 f L),MPV(>10.95 f L),were the independent risk factors of NEC.(2)Surfactant,probiotics befoe NEC and PDW(>11.8f L)were the protective factors of NEC.(3)The NEC prediction-scoring model is of some value in the prediction in the clinical.