| Objective To observe amplitude integratedelectroencephalo gram(aEEG)for early assessment of neurological function and neurodevelopmental outcome in high-risk neonates.Methods From January 2021 to January 2022,225 high-risk infants(including 5 infants with congenital brain abnormalities and genetic metabolic diseases),including 143 male infants,82 female infants,99 full-term infants,126 preterm infants,177 infants with brain injury group and 48 infants without brain injury group,were selected for inclusion in the study.177 cases,and 48 cases in the no brain injury group.They were further divided into 44 cases in group A from 27 to 31+6 weeks,40 cases in group B from 32 to 33+6 weeks,42 cases in group C from 34 to 36+6 weeks,and 42 of the 48 cases of fullterm high-risk infants without brain injury were selected as control group D.Bedside aEEG examination was completed within 72 hours of birth,aEEG results were graded,and cranial MR examination was completed at full gestational age and corrected gestational age at full term,and the DQ scores of the enrolled children at gestational age or corrected gestational age at 6 months were followed up.Statistical data were analyzed using SPSS 26.0 software,and ROC curves were plotted using Medcalc 15.0,which were statistically significant at P<0.05.The measurement data were described by median and quartile spacing[M(Q1,Q3)],and the Mann-Whitney U rank sum test was used for comparison between groups,and the count data were described by number of cases and composition ratio N(%),and the chi-square test was used for comparison between groups,and the Bonferroni method was used for two-way comparison.A binary logistic regression model was applied for analysis,and the model was evaluated by the ROC method.Results(1)General data:A total of 225 high-risk children were included,including 177 in the brain injury group and 48 in the no brain injury group.aEEG abnormalities were detected in 69.3%(156/225,including 115/126 in preterm infants and 45/99 in term infants),including a mild abnormality rate of 55.6%(125/225)and a severe abnormality rate of 13.8%(31/225);completed cranial MR in a total of 158,with an abnormality detection rate of 42.4%(67/158);5 cases lost to followup,5 cases died,and 24 cases with DQ scores ≤75,with an abnormality rate of 13.2%(29/220).The statistics of this study showed a significant increase in the odds of brain injury in other congenital disorders classified as many types and few cases were not analyzed in separate statistics,the top four of which were maternal cervical insufficiency(12 cases),convulsions(12 cases),intrauterine growth retardation(8 cases),congenital metabolic disorders and brain developmental malformations(5 cases).The combination of EEG and MRI examinations and DQ values suggested that cervical insufficiency,convulsions and congenital malformations and metabolic diseases can cause a higher chance of brain injury,while convulsions and congenital malformations and metabolic diseases can cause severe EEG changes and a high chance of poor neurological prognosis.(2)High risk factors for brain injury:gestational age(Z=-6.348,P<0.001),Arrhenius score(Z=-2.396,P=0.017)and birth weight(Z=-6.182,P<0.001)were lower in the group with brain injury than in the group with normal brain;the rate of other congenital diseases(χ2=12.193,P<0.001)was higher than in the group without brain injury,which was statistically significance.Gestational age and other congenital diseases were important influencing factors affecting brain injury.The risk of brain injury decreased with each week increase in gestational age(OR=0.697,95%CI:0.565-0.859);comparably for those without other congenital diseases,the risk of brain injury increased with other congenital diseases(OR=4.887,95%CI:1.677-14.237),which was statistically significant.(3)The predictive value of aEEG,cranial MR and the combination of the two in determining neurological injury in high-risk children:the sensitivity of aEEG diagnosis was 87.57%,the specificity was 97.92%,the AUC value was 0.927,the accuracy was 89.78%,the positive predictive value was 99.36%,and the negative predictive value was 68.12%;the sensitivity of cranial MR diagnosis was 36.16%,and the specificity was 93.75%.The sensitivity of the combined diagnosis was 93.22%,the specificity was 91.67%,the AUC value was 0.924,the accuracy was 92.89%,the positive predictive value was 97.63%,and the negative predictive value was 28.48%.was 78.57%.Analysis of variance showed that the three diagnostic modalities differed in AUC(χ2=6.056,P=0.048),sensitivity(χ2=27.548,P<0.001),accuracy(χ2=16.442,P=0.001)and negative predictive value(χ2=24.697,P<0.001),and the combined diagnosis of the two was statistically significant with the highest relative.(4)The change pattern of aEEG at different gestational ages:the normal rate of aEEG in the term infant group(χ2=60.814,P<0.001)was higher than that in the preterm infant group,which was statistically significant;the abnormal rate of Co(background continuity)(χ2=112.274,P<0.001)and SWC(χ2=115.667,P<0.001)were different among the four groups at different gestational ages.A two-by-two comparison showed that the abnormal rates of Co and SWC(sleep-wake cycle)in group D were lower than those in the other three groups,and the abnormal rates of Co and SWC in group C were lower than those in groups A and B,which were statistically significant.(5)The predictive value of aEEG,cranial MR and the combination of the two in determining neurological prognosis:the sensitivity of aEEG diagnosis was 93.10%,the specificity was 34.03%,the AUC value was 0.636,the accuracy was 41.81%,the positive predictive value was 17.64%,and the negative predictive value was 97.01%;the sensitivity of MR diagnosis was 41.38%,the specificity was 72.25%,AUC value of 0.568,accuracy of 68.18%,positive predictive value of 18.46%,and negative predictive value of 89.09%;the sensitivity of combined diagnosis of both was 93.10%,specificity of 34.03%,AUC value of 0.670,accuracy of 41.82%,positive predictive value of 17.65%,and negative predictive value of 97.01%.Analysis of variance showed that the sensitivity(χ2=23.824,P<0.001),specificity(χ2=20.629,P<0.001)and accuracy(χ2=8.895,P=0.012)of the three assessment methods were significantly different.Conclusion1,High risk factors for early neonatal brain injury are small for gestational age,low birth weight and low Apgar score;cervical insufficiency,convulsions and congenital malformations and metabolic diseases cause higher odds of brain injury.2.The Co and SWC abnormalities were higher in the preterm group than in the term group of high-risk neonates,and they increased with decreasing gestational age.3.aEEG combined with cranial MRI is more accurate than any single one of these tests for assessing early brain injury in high-risk neonates and poor neurological prognosis at 6 months of age(or 6 months of corrected gestational age). |
PDF Full Text Request