Intercenter variability and factors associated with survival without bronchopulmonary dysplasia in extremely preterm newborns.

Background: Variability in clinical practice may influence morbidity and mortality in extremely preterm infants. We aimed to know if there are differences in survival and survival without bronchopulmonary dysplasia (BPD) in extremely preterm infants in Spanish tertiary hospitals and the potential associated factors. Methods: Fifteen hospitals from the SEN1500 network were studied. The overall rate of survival without BPD was 61.4%. Hospitals with extreme results were grouped for comparison (Group 1; N  = 2480 versus Group 2; N  = 2367). A bivariate analysis of the characteristics of patients and perinatal interventions was carried out and the probability of survival without BPD was studied by Cox regression. Results: Survival (79.0 versus 72.9%; p  < .001) and survival without BPD (72.5 versus 49.1%; p  < .001) were greater in Group 1. Higher gestational age, birth weight, and female sex were associated with better outcomes. Oxygen administration (aHR: 0.868 [95%CI: 0.782, 0.964]; p =.008) and intubation (aHR: 0.767 [95%CI: 0.701, 0.839]; p  < .001) in delivery room were associated with lower survival without BPD. The occurrence of patent ductus arteriosus (PDA), sepsis and/or necrotizing enterocolitis (NEC) was independently associated with worse outcomes. After adjusting for confounders, the probabilities of survival without BPD were significantly higher among patients in Group 1: aHR: 1.557 [95% CI: 1.458, 1.662]; p  < .001. Conclusions: Among the studied hospitals, we found great variability in clinical practice and in the rates of survival and survival without BPD. A more conservative approach to the use of oxygen and respiratory support seems to be related to an increase in survival without BPD. Complications such as PDA, sepsis, and/or NEC decrease survival without BPD. Other variables not included in the present study could be relevant and deserve further study.