A Retrospective Multicenter Stratified Study on Perinatal Factors Influencing Neonatal Mortality in Preterm Infants in Kazakhstan.

Abstract

Aim: This study is aimed at evaluating the cumulative effect of postnatal risk factors on the survival of preterm neonates by examining key clinical parameters and complications across various gestational ages. Methods: A retrospective cohort study was conducted using data from 1109 neonates admitted to neonatal intensive care units at two tertiary regional hospitals in Kazakhstan between 2021 and 2024. Patients were classified into three groups based on gestational age: extremely preterm (< 28 weeks, n = 223), very preterm (28-31 weeks, n = 384), and moderate to late preterm (32-36 weeks, n = 502). Initially, to identify significant risk factors, categorical variables were analyzed using the χ ² test or Fisher's exact test with the Bonferroni correction, depending on whether the expected counts were ≥ 5 or < 5, while continuous variables were examined with the Kruskal-Wallis and Mann-Whitney U tests. Subsequently, multivariate logistic regression was applied to develop a prognostic model for each study group based on previously identified statistically significant risk factors for neonatal mortality. The predictive performance of these factors was further evaluated through ROC curve analysis with AUC. Finally, the Kaplan-Meier method was used to reflect overall mortality outcomes, illustrating their association with gestational age and the distribution of fatal cases over time in days. Results: The results of this study reveal significant differences in survival rates among preterm neonates based on gestational age, with mortality being most pronounced in infants born at less than 28 weeks of gestation. As gestational age increased among preterm infants, there was a marked reduction in the number of statistically significant independent risk factors identified in logistic models that influence survival. Disseminated intravascular coagulation consistently emerged as a significant predictor across all three logistic models. Neonatal depression, hyperbilirubinemia, and necrotizing enterocolitis were significant in both extremely preterm infants (less than 28 weeks) and very preterm infants (28-31 weeks). However, patent ductus arteriosus and bronchopulmonary dysplasia were statistically significant only in the extremely preterm group (less than 28 weeks). Conclusion: In a cohort of preterm neonates in Kazakhstan, the cumulative impact of various risk factors plays a critical role in determining survival outcomes, with increasing gestational age significantly enhancing the likelihood of survival. Further research is required to refine prognostic models and identify factors specific to different sociogeographical populations.