Send to

Choose Destination
PLoS One. 2019 Apr 4;14(4):e0214232. doi: 10.1371/journal.pone.0214232. eCollection 2019.

Early extubation is not associated with severe intraventricular hemorrhage in preterm infants born before 29 weeks of gestation. Results of an EPIPAGE-2 cohort study.

Author information

UMR 5525 ThEMAS, CNRS, TIMC-IMAG, Grenoble Alpes University, Grenoble, France.
Neonatal Intensive Care Unit, Grenoble Alpes University Hospital, Grenoble, France.
Obstetrical, Perinatal, and Pediatric Epidemiology ResearchTeam (Epopé), INSERM UMR 1153, Center for Epidemiology and Biostatistics Sorbonne Paris Cité; Paris Descartes University, Paris, France.
Risk in Pregnancy DHU, Cochin Hotel-Dieu Hospital, APHP, Paris, France.
EPIUnit-Institute of Public Health, University of Porto, Porto, Portugal.
Neonatal Intensive Care Unit, Jeanne de Flandre Hospital, CHRU Lille, Lille, France.
Neonatal Intensive Care Unit Port-Royal, Hôpital Cochin, APHP Paris, France.
Department of Neonatal Medicine, CHU Nantes, Nantes, France.
Department of Neonatal Medicine and Intensive care, CHU Rouen, Rouen, France.
Division of Neonatology and Pediatric Intensive Care, CHU Geneva, Geneva, Switzerland.



To determine whether there is an association between severe intraventricular hemorrhage and early extubation in preterm infants born before 29 weeks of gestational age and intubated at birth.


This study included 1587 preterm infants from a nationwide French population cohort (EPIPAGE-2). Secondary data on intubated preterm infants were analyzed. After gestational age and propensity score matching (1:1) we built two comparable groups: an early extubation group and a delayed extubation group. Each neonate in one group was paired with a neonate in the other group having the same propensity score and gestational age. Early extubation was defined as extubation within 48 hours of life. Severe intraventricular hemorrhages were defined as grade III or IV hemorrhages according to the Papile classification.


After matching, there were 398 neonates in each group. Using a generalized estimating equation model, we found that intraventricular hemorrhage was not associated with early extubation (adjusted OR 0.9, 95%CI 0.6-1.4). This result was supported by sensitivity analyses.


The practice of early extubation was not associated with an increased proportion of intraventricular hemorrhages. To complete these results, the long-term neurologic outcomes of these infants need to be assessed.

Conflict of interest statement

The authors have declared that no competing interests exist.

Supplemental Content

Full text links

Icon for Public Library of Science Icon for PubMed Central
Loading ...
Support Center