Abstract: Background：Electroencephalogram (EEG) monitoring has been widely used in the assessment of brain development in preterm infants and brain function in children with brain injury. There are few clinical reports on the changes of EEG in highrisk fullterm neonates without clinical symptoms in the early stage after birth. Objective：To investigate the necessity of early electroencephalogram (EEG) screening in asymptomatic highrisk neonates after birth. Design：Retrospective cohort study. Methods：The information at birth, maternal information, intrapartum information, discharge diagnosis, aEEG and vEEG results within 72 hours after birth of highrisk infants who underwent electrophysiological monitoring were retrospectively collected. vEEG was used as the gold standard for EEG monitoring, and aEEG was used as the screening method. Based on the five domains of language, gross motor, fine motor, problem solving and personalsocial in the Developmental Questionnaire (3), a selfmade neurobehavioral development screening questionnaire was made using the common similar questions in the BayleyScales of Infant Development at the age of 12 months. The outcome was neurobehavioral prognosis at 12 months of age. Main outcome measures：To evaluate the screening effect of early electroencephalogram (EEG) monitoring on neurobehavioral outcomes at 12 months of age. Results：A total of 398 highrisk infants who met the inclusion and exclusion criteria were enrolled in this study. There were 205 males (51.5%) and a gestational age of 38.9±1.1 (37+041+2) weeks, all of whom had stable vital signs, no hypoxia or nervous system clinical manifestations.aEEG and vEEG examinations were performed within 72h after birth. With vEEG as the gold standard and aEEG as the screening method, 46 cases were true positive, 325 cases were true negative, 0 case was false positive, and 27 cases were false negative. The sensitivity of aEEG was 63.0%(95%CI：51.7%74.4%), and the specificity was 100%(95%CI：100%100%). The vEEG abnormal group included 46 true positive cases and 27 false negative cases, and the vEEG normal group included 325 true negative cases and no false positive cases. A total of 386 cases (97.0%) completed the telephone followup of neurobehavioral questionnaire at 12 months of age. There was a significant difference in the total score of 8090 in the five domains of the Neurobehavioral Development Questionnaire between the normal vEEG group and the abnormal vEEG group (4.1% vs 40.3%).The scores of language, gross motor, fine motor, problem solving and personalsocial ability were significantly different. There were significant differences in birth weight, small for gestational age, length of hospital stay, vaginal delivery, forceps delivery, intrauterine distress, fetal heart rate <100 beats/min, umbilical artery blood gas pH value, BE value and lactic acid between the two groups. Conclusions：With vEEG as the gold standard, the sensitivity and specificity of aEEG were 63.0% and 100%, respectively. In total，4.1% and 40.3% of highrisk infants with normal and abnormal EEG had abnormal neurobehavioral development at 12 months of age, respectively. Therefore，aEEG screening is necessary for highrisk infants early after birth.

Key words: High-risk infants, Early electroencephalogram (EEG), Follow-up of neurobehavioral development