关键词:

极低出生体质量儿, 宫外生长发育迟缓, 预测模型, 危险因素, logistic 回归分析

Abstract:

Objective To observe the extrauterine growth and development of very low birth weight infants and establish a prediction model of extrauterine growth retardation（EUGR）. Methods A total of 804 very low birth weight infants admitted to our hospital from January 2019 to January 2021 were included as the research objects and were divided into EUGR group（n = 567）and non⁃EUGR group（n = 237）based on the differential diagnosis of EUGR. The basic information of the patients was compared between the two groups during hospitalization，and then the variables with differences were included in the logistic regression analysis. The EUGR was taken as the dependent variable（Y，yes = 1，no = 0），and other variables like intrauterine growth retardation（IUGR），birth weight，time to recover birth weight，time to start enteral nutrition，sepsis，NRDS and amino acid content at the end of the first week as the independent（X）. The risk factors of EUGR in the very low birth weight infants were analyzed，and finally the ROC curve of risk factors was drawn to establish the EUGR prediction model. Results Among the 804 very low birth weight infants，567 cases（70.52%）developed EUGR and they were all included in the EUGR group. Amongthem，the height of 254 cases was less than 2 standard deviations of the normal infants of the same age and sex at 6 months，the body weight of 285 cases had was less than 2 standard deviations of the normal infants of the same age and sex，and the head circumference of 154 cases was less than 2 standard deviations of the normal infants of same age and sex. There were 237 cases（29.48%）diagnosed with no EUGR and then assigned as the non⁃EUGR group. There were no significant differences in gender，gestational age，discharge weight，length of hospital stay，Apgar score，duration of parenteral nutrition，total oral feeding time，total oral maximum calorie，fat emulsion volume and energy intake at the first and second weekend，amino acid intake at the second weekend，purulent meningitis，BPD and ICH between the two groups（P > 0.05）. The birth weight and amino acid content at the end of the first week in the EUGR group were both significantly lower than those in the non⁃EUGR group，the start time of enteral nutrition and the recovery time of birth weight in the EUGR group were both significantly later than those in the non ⁃EUGR group，and the rates of the infants combined with IUGR，sepsis and NRDS were significantly higher than those in the non⁃EUGR group（P < 0.05）. Logistic regression analysis showed that IUGR，birth weight，time to recover to the birth weight，time to start enteral nutrition，sepsis，NRDS and amino acid content at the end of the first week were inde⁃ pendent influencing factors of EUGR in the very low birth weight infants（P < 0.05）. ROC analysis showed that birth weight，time to recover to the birth weight，time to start enteral nutrition，amino acid content at the end of the first week and area under the curve of combined prediction were 0.830，0.782，0.862，0.851 and 0.911，respectively. The area under the curve（0.911）and 95% IC（0.851⁃0.972）of combined prediction was better（P < 0.05）. Conclusion The incidence of EUGR in very low birth weight infants is high，and it has many influencing factors. Clinically，the formula of（Y=Xbirth weight ⁃2.134×X + 3.451 × X start enteral nutrition time + 4.160 ×+ X amount of amino acid at the end of the first week）is suggested to use in the prediction of EUGR in the very low birth weight infants. 

Key words: very low birth weight infants,  , extrauterine growth restriction,  , prediction model,  , risk factors,  , logistic regression analysis