Risk Factors of Neonatal Acute Respiratory Distress Syndrome Based on the Montreux Definition in Neonates with Sepsis: A Retrospective Case–Control Study

Objective The aim of the study is to analyze the risk factors for neonatal acute respiratory distress syndrome (NARDS) development based on the Montreux definition among near- and full-term neonates with sepsis and received meropenem. Study Design This was a single-center, case–control, retrospective trial from January 2019 to June 2020. Newborns of gestational ages (GAs) ≥35 weeks, diagnosed with sepsis and received meropenem were included. Patients who developed NARDS subsequently were defined as the study group (NARDS group), while the others without NARDS were enrolled in the control group (non-NARDS group). Results Out of 213 eligible neonates, NARDS occurred in 52 (24.4%) cases. In univariate analysis, infants with NARDS had a lower GA and birth weight, but a higher rate of premature birth (p <0.05). The median onset times of sepsis were earlier among neonates with NARDS compared with those without NARDS (1 [1,1] vs. 6 [1,15] days, p <0.001). Neonates with NARDS were more likely to suffer from early-onset sepsis (EOS), persistent pulmonary hypertension of newborns, pulmonary hemorrhage, septic shock, and patent ductus arteriosus (p <0.05). During labor, women whose neonates experienced NARDS were more likely to have a cesarean delivery (67.3 vs. 46.6%, p = 0.009) and likely to receive at least one dose of corticosteroids (21.2 vs. 5.0%, p = 0.001). In multivariable analyses, factors remaining independently associated with NARDS were premature birth, cesarean delivery, EOS, and septic shock. Compared with conventional inflammatory markers for NARDS, procalcitonin (PCT) was correlated with septic neonates who developed NARDS (p = 0.012) but had a low diagnostic value (area under the curve [AUC] = 0.609). C-reactive protein, white blood cells, and PLT did not correlate with morbidity of NARDS (AUC <0.05 and p >0.05). Conclusion Premature birth, cesarean delivery, EOS, and septic shock were independently associated with NARDS among near- and full-term septic neonates. PCT showed limited predictive value for NARDS. Key Points