Short-term neonatal in-hospital associations following delivery room ETI versus PPV resuscitation: a multicenter retrospective cohort study

Abstract Background Endotracheal intubation (ETI) and positive pressure ventilation (PPV) are core neonatal resuscitation interventions. Baseline clinical imbalances confound gestational-age (GA)-specific associations between ETI and in-hospital adverse events. We evaluated GA-stratified associations between resuscitation modes and these endpoints to guide individualized resuscitation; long-term follow-up data were unavailable and excluded herein. Methods This 2024 multicenter retrospective cohort analysis employed data collected via a pre-specified standardized case-report form. Eligible neonates received delivery room (DR) PPV resuscitation and were admitted to four Shenzhen tertiary Neonatal Intensive Care Units within 24 h post-birth with parental informed consent. Participants were stratified by GA and grouped by maximal DR resuscitation modality. Stratum-specific propensity score weighting (PrSW) was applied to balance baseline confounders. Stratified multiple binary logistic regression was used to quantify binary in-hospital adverse events and estimate crude and PrSW-adjusted ETI-endpoint associations. Results We enrolled 434 neonates, including 217 preterm infants (50.0%) and 217 full-term infants (50.0%). Among preterm infants, 148 (68.2%) received PPV and 69 (31.8%) received ETI; among full-term infants, 187 (86.2%) received PPV and 30 (13.8%) received ETI. Neonates who required ETI displayed poorer baseline profiles pre-adjustment. Stratum-specific weighting identified GA-dependent ETI–endpoint associations. After adjustment, ETI showed independent associations with multiple adverse events in preterm infants: pneumonia (aOR = 2.482, 95% CI 1.134–5.431, P = 0.023), pulmonary hemorrhage (aOR = 12.476, 95% CI 1.413–110.158, P = 0.023), respiratory distress syndrome (aOR = 6.156, 95% CI 2.06–18.397, P = 0.001), sepsis (aOR = 2.354, 95% CI 1.003–5.524, P = 0.049), and hyperglycemia (aOR = 4.075, 95% CI 1.808–9.186, P < 0.001). For full-term infants, ETI was independently associated with higher odds of sepsis (aOR = 12.773, 95% CI 2.739–59.573, P = 0.001), hyperglycemia (aOR = 46.268, 95% CI 9.067–236.114, P < 0.001), and composite adverse events (aOR = 8.899, 95% CI 2.743–28.875, P < 0.001). Most crude excess risks in the ETI group lost significance after weighting, with these gaps stemming from baseline imbalance. Conclusion ETI exhibits distinct associations with in-hospital adverse events across GA strata. Preterm ETI is associated with elevated odds of pulmonary, infectious and metabolic risks; full-term ETI is associated with higher odds of sepsis, hyperglycemia and composite adverse events. Standardized DR resuscitation protocols and targeted risk monitoring may reduce post-resuscitation morbidities. Long-term developmental events will be reported upon completion of follow-up.