Delivery Room Intubation Linked to Higher Odds of Newborn Complications in Multicenter Study

Every year, millions of newborns around the world need help taking their first breaths. For most, a mask and gentle bursts of positive pressure ventilation, known as PPV, are enough to inflate the lungs and start the transition to life outside the womb. For a smaller group of babies who do not respond, clinicians move to the more invasive option of endotracheal intubation, or ETI, in which a tube is threaded through the vocal cords directly into the trachea so that air and, when needed, surfactant can be delivered straight to the lungs. A new multicenter retrospective cohort study from Shenzhen, China, published in BMC Pediatrics, has now examined how these two resuscitation pathways relate to short-term in-hospital complications, and the findings suggest that the risks associated with intubation differ strikingly between preterm and full-term infants.

The research team, led by Hanni Lin of Shenzhen Luohu People’s Hospital and Jinan University together with colleagues across four Shenzhen tertiary neonatal intensive care units, analyzed data from 434 neonates who received delivery room PPV resuscitation and were admitted to one of the participating NICUs within 24 hours of birth during 2024. The cohort was deliberately balanced: exactly half of the infants, 217, were preterm, and half were full-term. Among the preterm babies, 148, or 68.2 percent, were managed with PPV alone, while 69, or 31.8 percent, required intubation. Full-term infants were far less likely to need the invasive route, with 187, or 86.2 percent, receiving PPV and only 30, or 13.8 percent, undergoing ETI. All families provided informed consent, and the study was approved by the Clinical Research Ethics Committee of Shenzhen Luohu People’s Hospital and conducted in accordance with the Declaration of Helsinki.

One of the central challenges in studying neonatal resuscitation is that babies who are intubated are, almost by definition, sicker to begin with. They tend to be born earlier, weigh less, and arrive with more compromised physiology. If researchers simply compare outcomes between intubated and non-intubated infants, the results can be badly distorted by these baseline differences, a problem known in epidemiology as confounding by indication. The Shenzhen team tackled this head-on using a statistical technique called propensity score weighting, abbreviated PrSW, which mathematically reweights the two groups so that measured baseline characteristics, such as gestational age, birth weight, antenatal steroid exposure, premature rupture of membranes, and mode of delivery, are balanced between them. The analysis was stratified by gestational age, meaning preterm and full-term infants were evaluated in separate strata, and stratified multiple binary logistic regression was then used to estimate both crude and weighted-adjusted odds ratios for a battery of in-hospital adverse events.

Before adjustment, the picture was stark. Neonates who required intubation displayed markedly poorer baseline profiles, and their crude rates of complications were correspondingly higher across nearly every endpoint the team measured, from respiratory distress syndrome to sepsis to metabolic disturbances. But the propensity-weighted analysis told a more nuanced story. Most of the crude excess risk in the ETI group lost statistical significance after weighting, indicating that a large share of the apparent harm reflected underlying illness rather than the procedure itself. Yet a set of associations survived the adjustment, and these residual, independently associated risks followed patterns that differed by gestational age in ways the authors describe as gestational-age dependent.

For preterm infants, intubation in the delivery room was independently associated with elevated odds of five distinct adverse outcomes. The adjusted odds ratio for pneumonia was 2.482, with a 95 percent confidence interval of 1.134 to 5.431 and a P value of 0.023. Pulmonary hemorrhage showed one of the strongest signals in the entire study, with an adjusted odds ratio of 12.476, though the confidence interval of 1.413 to 110.158 was wide, reflecting the small number of events. Respiratory distress syndrome carried an adjusted odds ratio of 6.156, with a confidence interval of 2.06 to 18.397 and a P value of 0.001. Sepsis was associated with an adjusted odds ratio of 2.354, and hyperglycemia with an adjusted odds ratio of 4.075, the latter with a P value below 0.001. Taken together, the preterm findings point to a cluster of pulmonary, infectious, and metabolic risks that travel with delivery room intubation even after accounting for how sick the babies were at baseline.

The full-term stratum produced a different and, in some respects, more surprising risk profile. In these infants, intubation was not significantly associated with the pulmonary complications that dominated the preterm analysis. Instead, the adjusted analysis revealed markedly elevated odds of sepsis, with an adjusted odds ratio of 12.773 and a confidence interval of 2.739 to 59.573, and an extraordinary association with hyperglycemia, whose adjusted odds ratio reached 46.268 with a confidence interval of 9.067 to 236.114 and a P value below 0.001. Full-term intubated infants were also far more likely to experience composite adverse events, with an adjusted odds ratio of 8.899. The authors suggest that because full-term infants rarely need intubation, those who do may represent an especially vulnerable subgroup, and the metabolic signal in particular may reflect the profound physiological stress of a difficult resuscitation in a baby whose lungs were otherwise mature.

The mechanistic logic behind these associations is plausible, though the study’s retrospective design means it can establish association rather than causation. Intubation bypasses the upper airway, which normally filters, warms, and humidifies inspired gas, and the procedure can introduce microtrauma and colonizing organisms into the trachea, pathways that could plausibly contribute to pneumonia and sepsis. Positive pressure delivered through a tube also generates different pressure distributions within the immature lung than mask ventilation, which may help explain the pulmonary hemorrhage and respiratory distress signals in preterm infants. Hyperglycemia, meanwhile, is a well-recognized stress response in critically ill neonates, and the strength of the metabolic associations in both strata suggests that the need for intubation may serve as a marker for a level of physiological derangement that conventional baseline variables do not fully capture.

The authors are careful about the limits of their work. As a retrospective cohort study relying on data collected through a pre-specified standardized case-report form, it cannot exclude residual confounding by unmeasured factors, such as the precise indication for intubation, the skill and experience of the resuscitating team, or the duration of the procedure. The single-year, single-city design, drawing on four tertiary NICUs within the Shenzhen Neonatal Data Network, may also limit generalizability to settings with different resources or case mixes. Long-term developmental outcomes were not available and were explicitly excluded from this analysis; the team states that follow-up data on longer-term events will be reported once that phase of the study is complete. The wide confidence intervals around some estimates, particularly for rare events like pulmonary hemorrhage, further underscore the need for cautious interpretation.

Even with those caveats, the study carries practical weight for a field in which delivery room decisions must be made in seconds. Current international resuscitation guidelines already emphasize that intubation should be reserved for infants who fail adequate PPV or who have specific indications, precisely because the procedure is technically demanding and carries its own risks. The new findings add gestational-age-specific texture to that principle: for preterm infants, the associated burden clusters in the lungs and in infection and metabolic control, while for full-term infants the dominant signals are sepsis, hyperglycemia, and overall composite morbidity. The authors argue that standardized delivery room resuscitation protocols and targeted post-resuscitation risk monitoring, tailored to gestational age, could help reduce these morbidities, for example by flagging intubated preterm infants for heightened surveillance of pulmonary complications and infection, and by monitoring glucose closely in any full-term infant who required a tube.

For clinicians and researchers alike, the study is a reminder that the first minutes of life cast long shadows across the entire hospital stay, and that the tools used to save a newborn’s breath are not clinically neutral. As neonatal care continues to improve survival at ever earlier gestations, disentangling the harms of underlying illness from the harms of the interventions themselves becomes both harder and more important. The Shenzhen team’s propensity-weighted, gestational-age-stratified approach offers a template for that disentangling, and their forthcoming long-term follow-up data will be watched closely by neonatologists eager to know whether the short-term signals documented here translate into lasting differences in neurodevelopment and health. In the meantime, the message for the delivery room is measured but clear: when it comes to intubating a newborn, less may often be more, and every tube placed deserves a plan for the risks that follow.

Subject of Research: Gestational-age-specific associations between delivery room endotracheal intubation versus positive pressure ventilation and short-term in-hospital adverse outcomes in neonates

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

Article References: Lin, H., Yu, Z., Tu, H., Zhang, A., Zhang, F., Li, G., Yuan, S., Fang, X., & Liu, J. (2026). Short-term neonatal in-hospital associations following delivery room ETI versus PPV resuscitation: a multicenter retrospective cohort study. BMC Pediatrics. https://doi.org/10.1186/s12887-026-07726-1

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07726-1

Keywords: neonatal resuscitation, endotracheal intubation, positive pressure ventilation, preterm infants, full-term infants, neonatal sepsis, hyperglycemia, respiratory distress syndrome, propensity score weighting, delivery room, NICU, retrospective cohort study