Establishment and validation of a prediction model for neonatal respiratory distress syndrome
Wang Yue , Ren Xiaochen , Wang Wenhui,et al.
1. Department of Pediatrics, Beijing Friendship Hospital, Capital Medical University, Beijing 100050, China; 2. Department of Pediatrics, Tangshan Maternal and Child Health Hospital, Hebei, Tangshan 063000, China
【Abstract】 Objective To investigate the incidence of neonatal respiratory distress syndrome (NRDS)
and its associated risk factors, and to develop and validate a nomogram prediction model for assessing the
risk of NRDS in newborns. Methods The clinical data of 200 newborns admitted to Beijing Friendship
Hospital, Capital Medical University from January 2020 to January 2023 were retrospectively analyzed,
based on the occurrence of NRDS, the newborns were divided into the unfavorable group (n=40, with
NRDS) and favorable group (n=160, without NRDS). Maternal and neonatal conditions during the follow-
up period were recorded and observed. Maternal data included age, history of adverse pregnancy outcomes, body weight, reproductive system diseases, mode of delivery, placental abruption, intrauterine infection, number of fetuses, gestational hypertension, and gestational diabetes mellitus. Neonatal data included gender, premature rupture of membranes, gestational age, amniotic fluid aspiration, body weight, intrauterine distress, meconium-stained amniotic fluid, birth asphyxia, and neonatal infection. Multivariate Logistic regression analysis was used to explore the influencing factors of NRDS in newborns. Receiver operating characteristic (ROC) curves and the area under the curve (AUC) were used to evaluate the predictive performance of each factor. Based on the above factors, a nomogram prediction model for the occurrence of NRDS in newborns was constructed, and consistency and discrimination were verified. Statistical analysis was performed by χ2-test. Results Among the 200 newborns, 40 cases developed NRDS, while 160 cases did not, with an incidence rate of 20.00% for NRDS in newborns. The proportions of intrauterine infection, multiple gestation, gestational hypertension, and gestational diabetes in the unfavorable group were higher than those in the favorable group [60.00% (24/40) vs 33.13% (53/160), 77.50% (31/40) vs 54.37% (87/160), 90.00% (36/40) vs 65.63% (105/160), and 77.50% (31/40) vs 41.25% (66/160), respectively; all P<0.05]. The proportions of amniotic fluid meconium staining, gestational age <37 weeks, birth weight <2 000 g, and birth asphyxia in the unfavorable group were higher than those in the favorable group [82.50% (33/40) vs 52.50% (84/160), 85.00% (34/40) vs 50.00% (80/160), 80.00% (32/40) vs 58.13% (93/160), and 87.50% (35/40) vs 46.88% (75/160), respectively; all P<0.05]. The results of multivariate Logistic regression analysis showed that multiple gestation gestational hypertension, gestational diabetes, neonatal birth weight <2 000 g, amniotic
fluid meconium staining, intrauterine infection, gestational age <37 weeks, and birth asphyxia were the
risk factors for the occurrence of NRDS in newborns (all P<0.05). The results of ROC curve analysis
showed that the above factors had good predictive value for the occurrence of NRDS in newborns (the
AUC values were 0.616, 0.747, 0.681, 0.609, 0.650, 0.634, 0.675, and 0.703 respectively; all P<0.05).
Based on the above factors, a nomogram prediction model for the occurrence of NRDS in newborns
was constructed. The AUC was 0.909, (95% CI : 0.847-0.970), and the C-index was 0.912. The
consistency and discrimination of this model both performed well. Conclusion Multiple gestation,
2026, Vol. 14 
